A Psychophysiological Mechanism Underlying Women's Weight-Management Goals: Women Desire and Strive for Greater Weight Loss Near Peak Fertility

Abstract

Three studies demonstrated that conception risk was associated with increased motivations to manage weight. Consistent with the rationale that this association is due to ovulatory processes, Studies 2 and 3 demonstrated that it was moderated by hormonal contraceptive (HC) use. Consistent with the rationale that this interactive effect should emerge when modern appearance-related concerns regarding weight are salient, Study 3 used a 14-day diary to demonstrate that the interactive effects of conception risk and HC use on daily motivations to restrict eating were further moderated by daily motivations to manage body attractiveness. Finally, providing evidence that this interactive effect has implications for real behavior, daily fluctuations in the desire to restrict eating predicted daily changes in women's self-reported eating behavior. These findings may help reconcile prior inconsistencies regarding the implications of ovulatory processes by illustrating that such implications can depend on the salience of broader social norms.
© 2015 by the Society for Personality and Social Psychology, Inc.

Full text

Personality and Social
Psychology Bulletin
2015, Vol. 41(7) 930 –942
© 2015 by the Society for Personality
and Social Psychology, Inc
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0146167215585726
pspb.sagepub.com
Article
Body weight is strongly tied to physical health. Not only is
having a high body weight associated with numerous health
problems, such as coronary heart disease, type 2 diabetes,
and high cholesterol (see Must et al., 1999), having a low
body weight is associated with other health problems, such
as malnutrition, poor muscle strength, and low bone density
(see Andreoli et al., 2001). Given that health-related goals
and intentions are a strong predictor of health-related behav-
iors (McEachan, Conner, Taylor, & Lawton, 2011), under-
standing the sources of people’s weight-management goals is
crucial to understanding and even promoting healthy weight
maintenance.
Two independent lines of research suggest a novel source
of weight-management goals among women. The first is a
body of research indicating that some women conform to
modern norms regarding attractive weight to secure and main-
tain interpersonal relationships. In many modern cultures, het-
erosexual women recognize that it is normative for men to
desire thin partners to heavier ones (e.g., Fallon & Rozin,
1985), and the salience of this norm varies across situations.
For example, women demonstrate elevated body-appearance
concerns when viewing various advertisements (Lavine,
Sweeney, & Wagner, 1999; Tiggemann & McGill, 2004) and
magazines (Morry & Staska, 2001), anticipating interacting
with a man (Calogero, 2004), and wearing a swimsuit (Hebl,
King, & Lin, 2004). Thus, although different women may
attempt to attract and secure partners in different ways (e.g.,
wearing make-up, tanning, using diet pills, through clothing
choice, etc.; see Buss, 1988; Durante, Griskevicius, Hill,
Perilloux, & Li, 2011; Haselton, Mortezaie, Pillsworth,
Bleske-Rechek, & Frederick, 2007; Hill & Durante, 2011),
some women attempt to attract and secure a partner by man-
aging their weight (Li, Smith, Griskevicius, Cason, & Bryan,
585726PSPXXX10.1177/0146167215585726Personality and Social Psychology BulletinMeltzer et al.
research-article2015
1Southern Methodist University, Dallas, TX, USA
2Florida State University, Tallahassee, USA
3University of Kentucky, Lexington, USA
4University of North Carolina at Greensboro, Greensboro, USA
Corresponding Author:
Andrea L. Meltzer, Southern Methodist University, P.O. Box 750442,
Dallas, TX 75275, USA.
Email: ameltzer@smu.edu
A Psychophysiological Mechanism
Underlying Women’s Weight-
Management Goals: Women Desire and
Strive for Greater Weight Loss Near Peak
Fertility
Andrea L. Meltzer1, James K. McNulty2, Saul L. Miller3,
and Levi R. Baker4
Abstract
Three studies demonstrated that conception risk was associated with increased motivations to manage weight. Consistent
with the rationale that this association is due to ovulatory processes, Studies 2 and 3 demonstrated that it was moderated
by hormonal contraceptive (HC) use. Consistent with the rationale that this interactive effect should emerge when modern
appearance-related concerns regarding weight are salient, Study 3 used a 14-day diary to demonstrate that the interactive
effects of conception risk and HC use on daily motivations to restrict eating were further moderated by daily motivations
to manage body attractiveness. Finally, providing evidence that this interactive effect has implications for real behavior,
daily fluctuations in the desire to restrict eating predicted daily changes in women’s self-reported eating behavior. These
findings may help reconcile prior inconsistencies regarding the implications of ovulatory processes by illustrating that such
implications can depend on the salience of broader social norms.
Keywords
ovulation, weight, health goals, relationships, women
Received August 20, 2014; revision accepted April 13, 2015
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 931
2010; Meltzer & McNulty, 2015; Meltzer, Novak, McNulty,
Butler, & Karney, 2013; Mori, Chaiken, & Pliner, 1987). In
one study, for example, women ate less in the presence of a
desirable man than when in the presence of a less desirable
man or another woman (Mori et al., 1987).
The second line of research is a growing literature demon-
strating that women’s goals to attract and secure relationship
partners shift across the ovulatory cycle. According to the
ovulatory-shift hypothesis (Gangestad & Thornhill, 2008),
women are increasingly motivated to attract partners with
features that were indicative of high genetic quality ances-
trally during peak fertility (i.e., approximately 12-16 days
before the next menstruation). These shifting motivations
lead to corresponding shifts in behavioral attempts to com-
pete for high-value mates (Cantú et al., 2014; Gangestad,
Thornhill, & Garver-Apgar, 2005; Haselton & Gangestad,
2006). Given the relative importance of physical appearance
to men (Li et al., 2013; Meltzer, McNulty, Jackson, & Karney,
2014), one way women near peak fertility compete for men
is to engage in a variety of behaviors that enhance their
appearance (Durante, Li, & Haselton, 2008; Haselton et al.,
2007; Hill & Durante, 2009). For example, women are more
likely to dress fashionably (Haselton et al., 2007) and choose
clothing that reveals more skin (Durante et al., 2008; Haselton
et al., 2007) when near peak fertility than when not.
Considering these two bodies of work together leads to
the novel hypothesis that women who tend to respond to
modern cultural norms regarding body attractiveness by
managing their weight may be particularly motivated to
manage their weight near peak fertility. Indeed, existing
physiological research on the hormonal correlates of eating
provides indirect support for this possibility. For example,
several studies demonstrate that estradiol, a hormone ele-
vated near peak fertility, is associated with restricted eating
in both humans and non-human mammals (Czaja & Goy,
1975; Edler, Lipson, & Keel, 2007). Furthermore, several
other studies demonstrate that women consume fewer calo-
ries near peak fertility (e.g., Brown, Morrison, Calibuso, &
Christiansen, 2008; Gong, Garrel, & Calloway, 1989; Saad
& Stenstrom, 2012; for an earlier review, see Fessler, 2003).
Nevertheless, this research provides only indirect support
for the current theoretical perspective that such behaviors
reflect women’s tendencies to respond to modern cultural
norms regarding body attractiveness. Specifically, none of
the existing studies demonstrate that ovulatory shifts in eat-
ing behavior are linked to psychological motivations and/or
appearance-related concerns. In fact, given that these effects
have emerged in both human and non-human animals, they
have been interpreted in terms of non-appearance-related
neuroendocrinological factors (Fessler, 2003; Frank, Kim,
Krzemien, & Van Vugt, 2010). Indeed, one explanation is
that this tendency is an adaption that allows for time-alloca-
tion trade-offs females make near peak fertility (i.e., time
devoted to activities related to reproduction versus time
devoted to activities related to food foraging; see Fessler,
2003). Although such trade-offs may indeed be one reason
for ovulatory shifts in eating among non-humans and humans
alike, it nevertheless remains possible that some human
females are additionally more motivated to lose weight near
peak fertility for body-appearance-related reasons. Providing
evidence for this psychological mechanism would deepen
our theoretical understanding of weight motivations by sug-
gesting a novel psychophysiological mechanism for such
effects.
Overview of the Current Research
We conducted three studies that examined whether women’s
weight-management goals shift across the ovulatory cycle.
Study 1 used a within-person design to examine whether
women desire to lose more weight when they are near versus
far from peak fertility. Study 2 aimed to replicate this impli-
cation of the ovulatory shift in an independent sample of
women and provide stronger evidence that it is indeed due to
ovulatory processes by examining whether it emerges only
among women who do not use hormonal contraceptives
(HCs; for example, the pill, vaginal ring, the patch; see
Durante et al., 2011; Fleischman, Navarrete, & Fessler,
2010). Finally, Study 3 examined whether the interactive
effect of fertility and HC use emerged only when women’s
desire to manage the appearance of their bodies was rela-
tively salient, as well as whether the fluctuating motivations
associated with ovulation predict a primary behavioral
mechanism of weight maintenance: caloric restriction. In
line with the notion that the predicted effect should be driven
by heterosexual women’s desire to manage the appearance of
their bodies for men, all three studies focused exclusively on
heterosexual women.
Study 1
Participants
Participants were 22 naturally cycling (i.e., not using HCs)
undergraduate women who were selected from a broader
within-person study because they were those who (a)
reported experiencing a regular menstrual cycle and (b) com-
pleted both components of the within-person design. This
broader study consisted of 39 heterosexual women (the total
number of participants was determined by the number of
undergraduates who volunteered for this study during a time
frame of one academic semester); however, 8 participants
failed to respond correctly to quality-control items and 7 par-
ticipants failed to complete both components of the within-
person design and thus could not be used in the within-person
analyses. Two additional participants were excluded from
analyses: 1 who was over the age of 35 (because women over
the age of 35 experience a significant decline in fecundabil-
ity; Rothman et al., 2013) and 1 who reported a desire to lose
an extreme amount of weight relative to the rest of the sample
by guest on June 9, 2015psp.sagepub.comDownloaded from

932 Personality and Social Psychology Bulletin 41(7)
(114.50 pounds, SD = 3.54, vs. the 16.64 pounds, SD = 15.10,
reported by the other 22 participants). Notably, with the
exception of the 1 participant who was excluded because she
reported a desire to lose an extreme amount of weight, the 22
women who were included in the analyses did not differ from
those who were excluded in either their current weight, t(37)
= 1.18, ns, or ideal weight, t(37) = −1.41, ns.1 These 22 par-
ticipants reported a mean age of 18.71 years (SD = 1.01);
most (81.80%) were Caucasian.
Procedures
Following approval from the Institutional Review Board, all
participants provided informed consent and then provided
information necessary for calculating their likelihood of con-
ception (see next section; Wilcox, Dunson, Weinberg,
Trussell, & Baird, 2001). To ensure within-person variability
in conception risk, we used this information to instruct par-
ticipants to complete online measures of ideal weight (and
measures of current weight and self-esteem, to be used as
covariates) on a high-fertility day and a low-fertility day.
High-fertility sessions were scheduled 16 days prior to their
next estimated menstruation (near estimated ovulation), and
low-fertility sessions were scheduled 6 days prior to their
next estimated menstruation, according to whichever day
came first (for similar procedures, see Miller & Maner,
2010). Sixteen participants completed their low-fertility ses-
sion first. Participants were compensated with course credit.
Materials
Conception risk. Participants varied in the exact day on which
they completed each session, and thus, we created a continu-
ous measure of conception risk. Consistent with prior psy-
chological research examining ovulation effects (e.g.,
Eastwick & Finkel, 2012; Haselton & Gangestad, 2006), par-
ticipants reported the start date of their previous menstrua-
tion and their average menstrual cycle length that was used
to (a) place women on a “standard” 29-day cycle and (b)
calculate conception risk (range = .000-.094) on the day of
participation using the reverse-cycle-day method (see
Garver-Apgar, Gangestad, & Thornhill, 2008) and actuarial
medical data (see Wilcox et al., 2001). Higher scores indi-
cate higher probability of conception with a single act of
unprotected intercourse. Given the variability in women’s
follicular phase compared with the luteal phase, this back-
ward-count method is preferred over the forward-count
method (Fehring, Schneider, & Raviele, 2006). Neverthe-
less, given recent concerns that researchers can choose
whichever method of estimation that provides support for
their predictions (Harris, Chabot, & Mickes, 2013), we
attempted to replicate our predicted effect with estimates
formed using the forward-cycle-day method (based only on
the self-reported start date of women’s previous menstrua-
tion; range = .000-.086; see Garver-Apgar et al., 2008).
Weight. Participants reported how much they would ideally
like to weigh, as well as their current height and weight, at
each session.
Covariate. Prior work demonstrates that women have lower
self-esteem when they are ovulating (Hill & Durante, 2009),
and such lower self-esteem may result in increased desired
weight loss (Striegel-Moore, Silberstein, & Rodin, 1986)
independent of ovulation. Thus, we assessed self-esteem at
each high- and low-fertility session using the Rosenberg
Self-Esteem Scale (Rosenberg, 1965) and controlled for it in
a supplemental analysis. This scale is a 10-item, Likert-type
scale on which participants responded from 1 = strongly dis-
agree to 4 = strongly agree. Higher scores indicate higher
global self-esteem. Internal consistency was high (α = .95 for
high-fertility sessions, α = .94 for low-fertility sessions).
Results
On average, participants weighed 149.91 (SD = 27.13)
pounds and wanted to ideally weigh 133.27 (SD = 15.32)
pounds, indicating that women wanted to lose 16.64 (SD =
15.10) pounds. Women were on average 65.34 (SD = 2.74)
inches tall and had an average body mass index (BMI) of
24.65 (SD = 3.93), which falls at the upper end of the normal
range as defined by the Centers for Disease Control and
Prevention.
To test the prediction that desired weight change varied
within women’s ovulatory cycle, we used multilevel model-
ing (using the Hierarchical Linear Modeling 6.08 program;
Bryk, Raudenbush, & Congdon, 2004) to estimate the
within-person association between fertility and desired
weight change by regressing participants’ ideal weight onto
the standardized score of person-centered (within-person
differences in) conception risk, controlling for the standard-
ized score of person-centered (within-person differences in)
current weight in the first level of the model, and the stan-
dardized scores of between-person means of conception risk
and current weight, as well as session order, in the second
level of the model. Removing the variance in ideal weight
that is shared with current weight leaves only the difference
between ideal weight and current weight (i.e., desired weight
change) to be explained by the remaining predictors.
Furthermore, statistically controlling for current weight in
this way allowed us to avoid confounding participants’
desired weight change with their current weight, a notable
problem in difference scores analyses (Edwards, 1994;
Griffin, Murray, & Gonzalez, 1999; Lord, 1967).
Results are reported in Table 1. Consistent with predic-
tions, within-person changes in conception risk were nega-
tively associated with desired weight; women desired greater
weight loss when they were closer to peak fertility than when
they were farther from peak fertility. Notably, participants’
current weight was marginally positively associated with
conception risk, b = 0.58, SE = 0.30, t(39) = 1.93, p = .061,
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 933
95% confidence interval (CI) = [−0.022, 1.192], effect size r
= .30, suggesting that women reported weighing more near
peak fertility. To ensure that the association between wom-
en’s ideal weight and conception risk emerged independent
of the association between women’s current weight and con-
ception risk, we regressed women’s conception risk onto
person-centered ideal weight and current weight, controlling
for between-person differences in ideal weight and current
weight in the second level of the model. Within-person
changes in ideal weight remained marginally negatively
associated with conception risk, b = 0.01, SE = 0.00, t(39) =
−1.77, p = .085, 95% CI = [−0.012, 0.001], effect size r = .27,
suggesting that changes in women’s current weight across
their ovulatory cycle did not account for changes in women’s
ideal weight across their ovulatory cycle. Notably, in this
small sample of 22 women, self-esteem was not associated
with within-person changes in conception risk, controlling
for session order, b = 0.02, SE = 0.02, t(39) = 1.21, ns, 95%
CI = [−0.016, 0.065], effect size r = .19. Furthermore, the
primary effect emerged as marginally significant controlling
for self-esteem, b = −0.41, SE = 0.21, t(36) = −1.94, p = .060,
95% CI = [−0.834, 0.012], effect size r = .31. Finally, three
additional supplemental analyses demonstrated that this pri-
mary within-person effect (a) replicated using the forward-
count estimate of conception risk, b = −0.49, SE = 0.23, t(36)
= −2.15, p = .038, 95% CI [−0.942, −0.035], effect size r =
.34; (b) held controlling for a dummy code of race, Caucasian
= 0, non-Caucasian = 1; b = −0.42, SE = 0.21, t(36) = −2.04,
p = .049, 95% CI = [−0.839, −0.008], effect size r = .32; and
(c) was not moderated by either within-person differences in
current weight, b = −0.42, SE = 1.78, t(36) = −0.24, ns, 95%
CI = [−3.988, 3.140], effect size r = .04, or between-person
differences in current weight, b = −0.23, SE = 0.28, t(36) =
−0.83, ns, 95% CI = [−0.788, 0.326], effect size r = .14.
Discussion
Study 1 provided preliminary support for the idea that ovula-
tion is associated with an increased desire to lose weight.
Consistent with predictions, naturally cycling, heterosexual
women were more likely to report a lower ideal weight,
controlling their current weight, during periods of higher
conception risk than lower conception risk.
Study 2
Nevertheless, Study 1 is limited in several ways. First, the
sample size was relatively small, and a number of participants
could not be included for reasons under their own control
(e.g., did not correctly respond to quality-control items, did
not complete both components of the within-person design).
If the effect size obtained in Study 1 was implausibly large,
these issues could have contributed to a spurious effect
(Funder et al., 2014). Second, it is possible that conception
risk was confounded with other within-person factors (unre-
lated to actual fertility) that predict weight-management goals
that may thus account for this association, such as physical
symptoms associated with time since previous menstruation
or time until next menstruation. One way to help rule out such
factors is by demonstrating that the association between con-
ception risk and weight is moderated by the use of HCs,
which suppress ovulation. Given these two limitations of
Study 1, Study 2 attempted to replicate the effect that emerged
in Study 1 but (a) relied on a larger sample size and (b) exam-
ined whether it was moderated by the use of HCs.
Participants
Participants were the 92 heterosexual, undergraduate women
participating in a broader study who provided usable data.
The original sample consisted of 98 women who provided
complete data (i.e., current weight, ideal weight, and the data
necessary to calculate conception risk; as in Study 1, the total
number of participants was determined by the number of
undergraduates who volunteered for this study during a time
frame of one academic semester); however, 4 participants
were excluded because they indicated they were either les-
bian or bisexual, 1 participant was excluded because she
indicated she was over the age of 35 (see Study 1), and, as in
Study 1, 1 participant was excluded because she reported
wanting to lose an extreme amount of weight relative to the
rest of the sample (105 pounds compared with the average
Table 1. Association Between Conception Risk and Ideal Weight in Study 1.
b SE 95% CI Effect size r
Intercept 133.96 1.53 [130.898, 137.027] —
Session order −0.82 1.66 [−4.134, 2.495] .12
Between-person current weight 12.40*** 1.38 [9.635, 15.167] .90
Between-person conception risk −2.31†1.23 [−4.770, 0.150] .41
Within-person current weight 1.06** 0.34 [0.382, 1.745] .46
Within-person conception risk −0.42* 0.21 [−0.839, −0.008] .32
Note. df = 18 for intercept, session order, and between-person effects; df = 37 for within-person effects. Effect size r = ttdf
22
/( )+. CI = confidence
interval.
†p < .10. *p < .05. **p < .01. ***p < .001.
by guest on June 9, 2015psp.sagepub.comDownloaded from

934 Personality and Social Psychology Bulletin 41(7)
11.45 pounds, SD = 12.85, reported by the other 92 partici-
pants). Notably, with the exception of the 1 participant who
was excluded because she reported a desire to lose an extreme
amount of weight, the 92 women who were included in the
analysis did not differ from those who were excluded in
either self-reported current weight, t(95) = 0.27, ns, or ideal
weight, t(95) = 0.67, ns.2 These 92 participants reported a
mean age of 19.34 years (SD = 1.59), and most (73.90%)
were Caucasian. Eighty participants reported that their aver-
age menstrual cycle length consistently fell within 26 to 32
days (i.e., was “regular”); 12 participants reported that their
average menstrual cycle length was “irregular.”
Procedures
In contrast to Study 1, which utilized a within-subjects
design, Study 2 utilized a between-subjects design in which
participants reported to the laboratory at random times
throughout their cycle to complete measures of current
weight, ideal weight, HC use, and conception risk. Thus, par-
ticipants in this study demonstrated a broader range of con-
ception risk values that span the entire ovulatory cycle
compared with those in Study 1. Participants were compen-
sated with course credit.
Materials
Conception risk. As in Study 1, women reported the start date
of their previous menstruation. But rather than report the
average length of their menstrual cycle, women in this study
reported whether their menstrual cycle was “regular” or
“irregular.” Thus, we used participants’ self-reported start
date of their previous menstruation to calculate their concep-
tion risk on the day of participation using the forward-
cycle-day method (see Garver-Apgar et al., 2008; Wilcox et
al., 2001) and actuarial medical data for women with regular
cycles (range = .000-.094) and irregular cycles (range = .000-
.065). Without more specific information about the average
length of these women’s cycles, we were unable to use the
reverse-cycle-day method to calculate conception risk.
Weight. As in Study 1, participants reported how much they
would ideally like to weigh and their current weight. Sixty-
eight participants also reported their height.
HC use. Participants indicated whether they used any form
of HC (e.g., the pill, patch, vaginal ring, etc.). HC use was
dummy coded, such that 0 = no HC use (n = 39) and 1 = HC
use (n = 53).
Results
On average, participants weighed 139.10 (SD = 28.20)
pounds and reported ideally wanting to weigh 127.65 (SD =
18.46) pounds, indicating that women wanted to lose 11.45
(SD = 12.85) pounds. Participants who reported their height
were 64.00 (SD = 2.75) inches tall and had an average BMI
of 24.30 (SD = 5.83), which, as in Study 1, falls at the upper
end of the normal range.
Although we were able to calculate a “conception risk
value” for all participants based on days since menstruation,
that value is only meaningful among women who do not use
HCs because HCs suppress ovulation. Thus, we predicted
that calculated conception risk values should interact with
HC use to only affect desired weight among women not
using HCs. To test this prediction, we regressed participants’
ideal weight onto their standardized score of conception risk,
HC use, and the Conception Risk × HC Use interaction, con-
trolling for their standardized score of current weight.
Results are reported in Table 2. As can be seen, consistent
with predictions, women’s HC use marginally moderated the
association between their desired weight change and concep-
tion risk. The interaction is depicted in Figure 1. Simple
slopes analyses demonstrated that whereas desired weight
change was not associated with conception risk among
women who reported using HCs, b = 0.08, SE = 0.80, t(87) =
0.10, ns, 95% CI = [−1.648, 1.812], effect size r = .01, desired
weight change was negatively associated with conception
risk among women who reported not using HCs, b = −2.56,
SE = 1.19, t(87) = −2.15, p = .034, 95% CI = [−4.936,
−0.184], effect size r = .23. In other words, consistent with
the idea that desired weight loss is associated with ovulation,
only naturally cycling women reported wanting to weigh less
near peak fertility. Additional analyses demonstrated that
participants’ current weight was not associated with concep-
tion risk, b = −1.38, SE = 2.97, t(90) = −0.47, ns, 95% CI =
[−7.318, 4.557], effect size r = .05, and that null effect was
not moderated by HC use, b = −3.78, SE = 6.32, t(88) =
−0.60, ns, 95% CI = [−16.416, 8.860], effect size r = .06.
Table 2. Associations Between Conception Risk, HC Use, and Ideal Weight in Study 2.
b SE 95% CI Effect size r
Intercept 127.71 1.07 [125.570, 129.842] —
Current weight 17.25*** 0.70 [15.852, 18.644] .94
CR −2.56** 1.19 [−4.936, −0.184] .23
HC use −0.26 1.41 [−3.074, 2.550] .02
CR × HC use 2.64†1.47 [−0.300, 5.584] .19
Note. df = 87. Effect size r = ttdf
22
()/.
+ HC = hormonal contraceptive; CI = confidence interval; CR = conception risk.
†p < .10. **p < .01. ***p < .001.
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 935
Furthermore, two additional analyses demonstrated that this
Conception Risk × HC Use interaction (a) remained margin-
ally significant controlling for a dummy code of race,
Caucasian = 0, non-Caucasian = 1; b = 2.44, SE = 1.45, t(86)
= 1.68, p = .096, 95% CI = [−0.463, 5.349], effect size r =
.18, and (b) was not further moderated by between-person
differences in current weight, b = 0.46, SE = 2.09, t(84) =
0.22, ns, 95% CI = [−3.734, 4.654], effect size r = .02.
Discussion
Study 2 provided additional evidence that ovulation is asso-
ciated with an increase in women’s motivation to manage
their weight. Not only did Study 2 replicate the association
that emerged in Study 1 by demonstrating that conception
risk was associated with the desire to lose more weight
among naturally cycling women, Study 2 also provided
stronger evidence that this association was due to ovulation
by showing that it depended on HC usage and did not emerge
among women who reported using HCs.
Study 3
Although Studies 1 and 2 provide convergent evidence for
the role of ovulation in shaping women’s desires to be thin,
several important questions remain. First, like the large
majority of research examining the link between human ovu-
lation and mating motivations (see Gildersleeve, Haselton, &
Fales, 2014), Studies 1 and 2 were conducted on college
undergraduate students. Given that ovulation is theorized to
be particularly important to women’s short-term mating psy-
chology, it may not have the same implications for women in
more committed relationships like marriage. Indeed, one
recent study (Durante, Rae, & Griskevicius, 2013) demon-
strates that ovulation had very different implications for
women’s voting preferences depending on whether those
women were single or in committed relationships. It is thus
possible that ovulation is differentially related to women’s
weight motivation in marriage. This is particularly impor-
tant, given that marriage appears to be a significant source of
weight gain (Jeffery & Rick, 2002). Nevertheless, given that
even some married women may be motivated to appear
attractive by managing the appearance of their bodies near
peak fertility, whether for their partners or for other men,
ovulation may be associated with an increased desire to be
thin even among such women. Thus, Study 3 attempted to
replicate the effect in a sample of married women. Second,
although consistent with the theoretical framework that fer-
tility shifts lead women to desire to manage their weight
when norms regarding body weight and thus the desire to
enhance the appearance of their bodies are salient, Studies 1
and 2 did not provide evidence that ovulation was associated
with the desire to lose weight for body-appearance-enhanc-
ing reasons. As noted earlier, only some women attempt to
manage their weight to attract and/or secure relationship
partners, and the current theoretical framework suggests, it is
these women who should demonstrate an increased desire to
manage their weight near peak fertility. That is, although het-
erosexual women are more motivated to appear attractive to
men at peak fertility on average, only women who respond to
such motivations by improving or maintaining the appear-
ance of their bodies should desire to manage their weight;
other women may respond to ovulation by managing non-
body-related aspects of their appearance (e.g., wearing
make-up, tanning, through clothing choice, etc.). To provide
necessary support for the idea that the way in which the fer-
tility-related motivations to appear attractive manifest
depends on which attractiveness-related norms are salient at
a given time, Study 3 examined whether this effect was
stronger among naturally cycling women who were particu-
larly motivated to enhance the appearance of their bodies.
Finally, Studies 1 and 2 did not demonstrate that shifting
fluctuations in women’s weight-maintenance goals have any
implications for women’s actual behavior. Although a robust
literature indicates that people’s intentions and motivations
are one of the strongest predictors of behaviors (e.g.,
McEachan et al., 2011), situational factors can constrain
these links (Fazio, 1990), and it is possible that ovulation is
associated with various situational factors that constrain the
implications of women’s weight motivations for their actual
behaviors. Thus, Study 3 examined whether the association
between women’s weight motivations and fertility shifts
translated into actual eating behaviors.
Participants
Participants were the 89 married women participating in a
broader study of marital couples who provided the informa-
tion necessary to compute their conception risk.3 This
broader study consisted of 131 heterosexual couples (the
total number of participants was determined by the maxi-
mum number of couples we were able to compensate for
118
120
122
124
126
128
130
132
HC Use No HC Use
Ideal Weight
Low CR
High CR
Figure 1. Interactive effects of conception risk (CR) and
hormonal contraceptive (HC) use on women’s ideal weight,
controlling current weight, in Study 2.
Note. Low conception risk = .000; high conception risk = .094. We used
the lowest and highest values of conception risk because they represent
low and peak fertility in women’s ovulatory cycles.
by guest on June 9, 2015psp.sagepub.comDownloaded from

936 Personality and Social Psychology Bulletin 41(7)
participation); however, 6 wives were excluded because they
were older than 35 (see Studies 1 and 2), 1 wife was excluded
because she did not complete the daily diary portion of the
study, and an additional 35 wives were excluded because
they did not provide information necessary for computing
conception risk (e.g., were pregnant, had experienced meno-
pause, or reported that their most recent menstrual period
occurred over 60 days prior for other reasons). The 89 women
who were included in the analyses did not differ from those
who were excluded in their current weight, t(128) = 0.78, ns;
mean motivation to restrict eating, t(129) = −0.65, ns; mean
motivation to manage body attractiveness, t(129) = −1.28,
ns; or mean tendency to actually restrict eating, t(129) =
−0.46, ns (2 wives did not report current weight).4
Couples were recruited through (a) fliers around a large
university in the Southeastern United States, (b) craigslist.
com, and (c) facebook.com. On average, these 89 wives were
29.09 (SD = 3.42) years of age (2 wives did not report their
age), had been married for 4.22 (SD = 3.30) years, and earned
US$34,848.17 (SD = 28,636.96) per year; 85.4% self-identi-
fied as “White or Caucasian,” 4.5% self-identified as
“Asian,” 4.5% self-identified as “Hispanic or Latina,” 2.2%
self-identified as “Black or African American,” and 3.4%
self-identified as “two or more ethnicities.”
Procedures
Couples who responded to any of the methods of solicitation
were emailed a link to an initial online battery of question-
naires. These questionnaires included items necessary for
calculating wives’ conception risk, an item assessing wives’
HC use, an item assessing wives’ current weight and a mea-
sure assessing wives’ self-esteem (each to be used as covari-
ates), and various questionnaires beyond the scope of the
current analyses. Subsequent to completing these question-
naires, each member of the couple was emailed a link to a
daily questionnaire every day for 14 days. For wives, this
daily questionnaire included items assessing weight motiva-
tions, appearance motivations, and dieting behaviors. After
the 14th day, couples were mailed a check for participating.
Couples were paid US$15 per person for completing the
baseline questionnaires and US$1 per person per diary day
completed. As an incentive to complete more diaries, each
person was additionally paid US$11 for completing at least
12 diaries if they also completed a follow-up survey 6 months
later, which is beyond the scope of the current analyses.
Wives completed an average of 13.42 (SD = 1.17) days;
96.6% of wives completed at least 12 days.
Measures
Daily conception risk. Given that the wives in this study pro-
vided their average menstrual cycle length at baseline, we
estimated wives’ conception risk for each day of the daily
diary using the same reverse-cycle-day method used in Study
1. Specifically, we used wives’ self-reported start date of
their previous menstruation and average menstrual cycle
length to place all wives on a 29-day cycle and calculate their
conception risk for each day of the 14-day diary (see Garver-
Apgar et al., 2008; Wilcox et al., 2001). Notably, also as in
Study 1, we attempted to replicate any effects with estimates
formed using the forward-cycle-day method to calculate
women’s conception risk.
HC use. As in Study 2, wives indicated whether they used
HCs. HC use was dummy coded, such that 0 = no HC use
(n = 46) and 1 = HC use (n = 43).
Daily motivation to restrict eating. We assessed wives’ motiva-
tion to restrict eating each day of the daily diary using one
item: “To what extent were you motivated to restrict how
many calories you ate today?” where 1 = not at all and 7 =
very. In the interest of full disclosure, we also assessed wives’
motivation to lose weight each day of the daily diary using
one item: “To what extent were you motivated to lose weight
today?” where 1 = not at all and 7 = very. Conception risk,
HC use, and daily motivation to manage body attractiveness
did not significantly interact to account for variance in wom-
en’s daily motivation to lose weight, b = −0.04, SE = 0.08,
t(86) = −0.74, ns, 95% CI = [−0.198, 0.122], effect size
r = .08.5
Daily motivation to manage body attractiveness. We assessed
wives’ motivation to manage the appearance of their bodies
each day of the daily diary using one item: “To what extent
were you motivated to improve/maintain the appearance of
your body today?” where 1 = not at all and 7 = very.
Daily eating behavior. We assessed whether wives restricted
their eating each day using one item: “Did you restrict how
many calories you ate today?” where 0 = no and 1 = yes.
Covariates. As in Studies 1 and 2, participants reported their
current weight, which we used as a covariate in all analyses,
and height at baseline. On average, participants weighed
157.78 (SD = 42.90) pounds, were 66.34 (SD = 2.49) inches
tall, and had an average BMI of 25.06 (SD = 6.41), which
falls in the overweight range. In addition, as in Study 1, par-
ticipants reported their self-esteem (assessed at baseline with
the Rosenberg [1965] Self-Esteem Scale, α = .92), which
was used as a covariate in a supplemental analysis.
Results
In line with the idea that the association between women’s
conception risk and desired weight loss should be particu-
larly strong among women who were (a) not taking HCs (as
in Study 2) and (b) motivated to have an attractive body, we
tested whether the association between conception risk and
the motivation to restrict eating was moderated by (a) HC
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 937
use and (b) women’s motivation to manage the appearance of
their body. Specifically, we regressed women’s daily reports
of their motivation to restrict their eating onto the standard-
ized scores of their daily conception risk, daily motivation to
manage their body appearance, and the Conception Risk ×
Body Appearance Motivation interaction in the first level of
a multilevel model (using the Hierarchical Linear Modeling
6.08 program), and then entered HC use, which did not vary
across the days, onto all necessary Level-2 equations to esti-
mate the crucial Conception Risk × HC Use × Body
Appearance Motivation interaction, along with all lower-
order effects, controlling for day of assessment, the standard-
ized score of current weight, and between-person differences
in the motivations to restrict eating on the Level-2 intercept.
The results are reported in Table 3. As can be seen, consis-
tent with Studies 1 and 2, conception risk was positively
associated with daily motivations to restrict eating and, con-
sistent with Study 2, this association was moderated by HC
use. Nevertheless, consistent with predictions, this interac-
tion was further moderated by the motivation to manage
body attractiveness. To decompose this interaction, we first
examined the Conception Risk × HC Use interactions sepa-
rately for those high (1 SD above the sample mean) versus
low (1 SD below the sample mean) in motivations to manage
the appearance of their body. Among women low in body-
appearance motivations, the Conception Risk × HC Use
interaction was not significant, b = 0.01, SE = 0.09, t(86) =
0.18, ns, 95% CI = [−0.156, 0.186], effect size r = .02, and
conception risk was not associated with motivation to restrict
eating on average, b = −0.02, SE = 0.07, t(86) = −0.35, ns,
95% CI = [−0.157, 0.111], effect size r = .04. Among women
high in body-appearance motivations, in contrast, the
Conception Risk × HC Use interaction was significant, b =
−0.31, SE = 0.12, t(86) = −2.58, p = .012, 95% CI = [−0.547,
−0.069], effect size r = .27. This two-way interaction is
depicted in Figure 2. We further decomposed this interaction
by estimating the simple effects of high body-appearance-
motivated women’s conception risk for those using versus
not using HCs. Consistent with predictions, high body-
appearance-motivated women’s conception risk was posi-
tively associated with the motivation to restrict eating among
those not using HCs, b = 0.25, SE = 0.09, t(86) = 2.80, p =
.006, 95% CI = [0.072, 0.432], effect size r = .29, but unre-
lated to the motivation to restrict eating among those using
HCs, b = −0.06, SE = 0.08, t(86) = −0.70, ns, 95% CI =
[−0.214, 0.102], effect size r = .08. Notably, self-esteem was
not significantly associated with between-person differences
in conception risk on average, b = 0.3, SE = 0.06, t(86) =
0.54, ns, 95% CI = [−0.087, 0.150], effect size r = .06.
Although this null effect was moderated by HC use, b = 0.26,
SE = 0.12, t(84) = 2.20, p = .031, 95% CI = [0.023, 0.490],
effect size r = .23, the negative association between concep-
tion risk and self-esteem among those using HCs did not
Table 3. Associations Between Conception Risk, HC Use, Attractiveness Motivations, and Motivations to Restrict Eating in Study 3.
b SE 95% CI Effect size r
Intercept 3.12 0.08 [2.959, 3.271] —
Current weight 0.08†0.04 [−0.004, 0.155] .20
Mean eating motivations 0.90*** 0.07 [0.756, 1.052] .80
Day −0.01 0.01 [−0.033, 0.011] .11
Within-person CR 0.11* 0.06 [0.003, 0.226] .22
HC Use −0.16 0.11 [−0.384, 0.057] .16
BAM 0.92*** 0.09 [0.738, 1.102] .74
CR × HC use −0.15†0.08 [−0.301, 0.009] .20
CR × BAM 0.14* 0.06 [0.025, 0.251] .25
HC use × BAM −0.17 0.11 [−0.402, 0.055] .16
CR × HC use × BAM −0.16* 0.07 [−0.300, −0.023] .24
Note. df = 87 for day; df = 84 for intercept, current weight, mean eating motivations, and HC use; df = 86 for all other effects. Effect size r = ttdf
22
()/.
+
HC = hormonal contraceptive; CI = confidence interval; CR = conception risk; BAM = body attractiveness motivations.
†p < .10. *p < .05. ***p < .001.
3.2
3.4
3.6
3.8
4
4.2
4.4
4.6
4.8
HC UseNo HC Use
Motivation to Restrict Eating
Low CR
High CR
Figure 2. Interactive effects of conception risk (CR) and
hormonal contraceptive (HC) use on women’s motivation
to restrict eating among women high in body attractiveness
motivations, controlling current weight, in Study 3.
Note. Low conception risk = .000; high conception risk = .094. We used
the lowest and highest values of conception risk because they represent
low and peak fertility in women’s ovulatory cycles.
by guest on June 9, 2015psp.sagepub.comDownloaded from

938 Personality and Social Psychology Bulletin 41(7)
reach significance, b = −0.09, SE = 0.08, t(84) = −1.13, ns,
95% CI = [−0.255, 0.071], effect size r = .12. Furthermore, a
supplemental analysis demonstrated that this effect held con-
trolling for self-esteem, b = −0.17, SE = 0.07, t(86) = −2.48,
p = .015, 95% CI = [−0.300, −0.032], effect size r = .26. In
addition, current weight at baseline was not associated with
between-person conception risk at baseline, b = −5.34, SE =
4.63, t(85) = −1.16, ns, 95% CI = [−14.599, 3.911], effect
size r = .12, and this null effect was not moderated by HC
use, b = 2.21, SE = 9.23, t(83) = 0.24, ns, 95% CI = [−16.246,
20.668], effect size r = .03. Finally, additional supplemental
analyses demonstrated that the Conception Risk × HC Use ×
Body Appearance Motivation interaction (a) replicated using
the forward-count estimate of conception risk, b = −0.21, SE =
0.08, t(86) = −2.75, p = .007, 95% CI = [−0.367, −0.058],
effect size r = .28, (b) held controlling for race, where
Caucasian = 0 and non-Caucasian = 1, b = −0.16, SE = 0.07,
t(86) = −2.26, p = .027, 95% CI = [−0.297, −0.018], effect size
r = .24, and (c) was not further moderated by between-person
differences in current weight, b = −0.06, SE = 0.07, t(84) =
−0.76, ns, 95% CI = [−0.206, 0.092], effect size r = .08.
Next, we tested whether this Conception Risk × HC Use
× Body Appearance Motivation interaction indirectly pre-
dicted women’s actual eating behavior through women’s
motivation to restrict their eating by computing an asymmet-
ric CI for the indirect effect on eating behavior using the
RMediation program and procedure developed by Tofighi
and MacKinnon (2011). This procedure requires two steps.
The first is to estimate the effect of the distal predictor, in this
case the Conception Risk × HC Use × Body Appearance
Motivation interaction, on the expected mediator, in this case
women’s motivations to restrict their daily caloric intake.
This association is the one tested and observed in the previ-
ous paragraph. The second step involved estimating the
effect of the predicted mediator, in this case, women’s moti-
vations to restrict their daily caloric intake, onto the predicted
outcome, in this case, women’s daily eating behavior, con-
trolling for the distal predictor. Thus, we repeated the analy-
sis described above except this time we substituted women’s
daily eating behavior for women’s daily eating motivations
as the dependent variable and added daily eating motivations
as a time varying predictor in Level 1 of the model, with a
random effect. Importantly, given that any indirect, interac-
tive effects of conception risk on eating behavior should be
driven by within-person fluctuations in eating motivation,
rather than any between-person differences in eating motiva-
tion, we person-centered eating motivations and entered the
between-person mean differences in eating motivation on the
Level-2 intercept. Given the binary nature of the dependent
variable, we specified a Bernoulli sampling distribution.
Consistent with the second criterion necessary to establish
mediation, this analysis indicated that women’s motivations to
restrict their daily caloric intake significantly positively pre-
dicted their daily eating behavior, controlling for conception
risk, HC use, body-attractiveness motivations, all possible
interactions, day of report, and baseline current weight, b =
0.74, SE = 0.05, t(87) = 14.76, p < .001, 95% CI = [0.637,
0.837], effect size r = .85. Finally, we multiplied the estimate
of that effect together with the estimate of the Conception
Risk × HC Use × Body Appearance Motivation interaction to
obtain an estimate of the indirect effect, b = −0.12, and cal-
culated the 95% CI = [−0.223, −0.017] that indicated the
indirect effect was significant. In other words, women who
reported not using HCs were more likely to restrict their eat-
ing on days when they were more fertile and oriented toward
managing the appearance of their bodies because such
women were more motivated to do so.
General Discussion
Just as heterosexual women’s interpersonal goals shift across
their ovulatory cycle (Gangestad & Thornhill, 2008), so does
one of their important health-related goals. Across three
independent studies, women demonstrated higher motiva-
tions to lose weight near peak fertility. Consistent with the
idea that this tendency is driven by physiological changes
associated with ovulation, it was moderated by HC usage in
Studies 2 and 3, such that only women who did not use HCs
demonstrated higher motivations to lose weight near peak
fertility. Furthermore, consistent with the idea that this effect
was at least partially driven by women’s desire to have an
attractive body, it only emerged among women who reported
high motivations to manage the appearance of their body.
Finally, consistent with the idea that even small fluctuations
in motivations have strong effects on behavior (McEachan
et al., 2011), within-person changes in motivations to restrict
their caloric intake were strongly positively associated with
their actual restricted eating behavior in Study 3. Although
the sample size and resultant power of Study 1 was rather
modest, the fact that the key effect replicated in two addi-
tional larger studies assuages concerns that the significant
effect that emerged in Study 1 was due to an implausibly
large effect size. Indeed, the effect size of the key simple
effect between conception risk and ideal weight/motivation
to restrict eating among women not using HCs and motivated
to manage the appearance of their bodies (Study 3) was quite
consistent across the three studies (in Study 1, effect size r =
.32; in Study 2, effect size r = .23; in Study 3, effect size r =
.29). Furthermore, the chance that the predicted effect
emerged by chance in all three studies is very small
(Murayama, Pekrun, & Fiedler, 2014). In addition, not only
did the primary effect replicate across three independent
studies, it (a) emerged in two samples of undergraduate
women from different universities and a sample of married
women, (b) did not vary across participants’ weight, using
two samples of women who had a normal weight on average
and one sample of women who were overweight on average,
(c) utilized two different methods of assessing conception
risk (i.e., the reverse- and forward-cycle-day methods), (d)
provided evidence for the theoretical mechanism, and (e)
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 939
held controlling for women’s actual body size and race in all
studies and self-esteem in Studies 1 and 3.
Implications and Future Directions
These findings have several important implications. First,
they join an emerging literature highlighting the important
role of interpersonal goals in health-related processes (e.g.,
Pietromonaco, Uchino, & Dunkel Schetter, 2013). What is
unique about these findings, however, is that they demon-
strate the interactive effects of two different levels of such
interpersonal motivations: biological and socio-cultural.
Biological processes associated with shifting levels of hor-
mones across the cycle interacted with women’s desires to
conform to socio-cultural norms regarding body attractive-
ness to predict an important behavior: eating. Future research
may benefit by considering other ways in which these two
levels of interpersonal motivation jointly affect important
behaviors.
Second, these findings add to the growing literature dem-
onstrating that women, and many non-human mammals,
consume fewer calories near peak fertility (e.g., Brown et al.,
2008; Fessler, 2003). Previous research has explained this
phenomenon in terms of a time-allocation trade-off between
mating and foraging (Fessler, 2003). To be clear, the current
findings regarding weight motivations do not undermine
these explanations. Instead, they suggest an additional rea-
son why women may consume fewer calories near peak fer-
tility: Increased motivations to be attractive interact with
cultural standards regarding body attractiveness to make
some women manage the attractiveness of their bodies.
Future research may benefit by exploring the extent to which
these various mechanisms operate, either independently or
synergistically, to shape eating patterns across women’s
menstrual cycles. Indeed, both appear to be driven by pro-
cesses that vary across the ovulatory cycle (e.g., estradiol).
Third, the current studies also join a growing body of
research (e.g., Cantú et al., 2014; Durante et al., 2013) in
highlighting the importance of contextual factors in psycho-
logical research on ovulation. Some recent research demon-
strates that contextual factors interact with ovulation to
predict women’s behavior—for example, fertile women
engage in more flirtatious behavior in the presence of attrac-
tive and dominant men than when in the presence of less
attractive, less dominant men (Cantú et al., 2014). Although
Studies 1 and 2 of the current research demonstrated a main
effect of ovulation on women’s desired weight loss among
women not using HCs, Study 3 demonstrated that this effect
was moderated by women’s motivations to manage the
appearance of their body, which tend to be heightened by
modern standards regarding what is attractive (Meltzer &
McNulty, 2015). Future research may similarly benefit by
considering the broader context in which women are situated
to best understand the implications of ovulation for women’s
mating and health psychology. Indeed, failing to recognize
and consider such nuances may help explain controversies
over the inconsistencies that have emerged in ovulation
research (see Gildersleeve et al., 2014; Harris et al., 2013).
Although these findings may have limited direct implica-
tions for changes in women’s weight, because actual weight
management requires motivations and behaviors that span
more than several (fertile) days, the general theoretical
notion that socio-cultural processes interact with biological
processes to predict behavior likely has important broader
practical implications. For example, women who are moti-
vated to manage the appearance of their bodies may be more
likely to take diet pills, which tend to have numerous harmful
side effects even when only enacted occasionally (Kernan
et al., 2000; Pittler, Schmidt, & Ernst, 2005), when they are
closer to ovulation. Likewise, because tanned skin is per-
ceived as attractive in Western culture (Hillhouse, Stair, &
Adler, 1996), such women may be more likely to engage in
sun (or sunless) tanning near peak fertility—behaviors that
are also linked to health problems even when only enacted
occasionally (Boniol, Autier, Boyle, & Gandini, 2012).
Indeed, naturally cycling women report that they tan more
frequently near peak fertility than during less-fertile phases
of their cycle (Saad & Stenstrom, 2012). Future research
may benefit by examining the extent to which these and
other risky behaviors vary across women’s ovulatory cycles.
Strengths and Limitations
Several strengths of the current research enhance our confi-
dence in the results reported here. First, as noted earlier, the
ovulatory shift of desired weight loss replicated across three
independent studies, helping to alleviate concerns regarding
the use of a small sample in Study 1 and provides confidence
in the reproducibility of the primary effect (Funder et al.,
2014; Murayama et al., 2014). Second, Study 1 demonstrated
the within-person effect, helping to rule out between-person
confounds. Third, providing evidence that the primary effect
emerged due to hormonal fluctuations across the cycle, it
was marginally moderated by HC use in Study 2 and signifi-
cantly moderated by HC use in Study 3. Fourth, providing
support for the theoretical framework that guided our predic-
tion, Study 3 additionally demonstrated that the primary
effect emerged only among naturally cycling women who
were highly motivated to manage the appearance of their
bodies. Finally, as also noted earlier, analyses in the current
studies controlled several potential confounds (e.g., current
weight, race, self-esteem), thus decreasing the possibility
that the results were due to associations with those
variables.
Nevertheless, several factors limit interpretations of the
current findings until they can be extended. First, although
estimations of conception risk have been successfully used in
prior research (e.g., Haselton & Gangestad, 2006; Miller &
Maner, 2010), some participants may have nevertheless inac-
curately reported the start of their previous menstruation and/
by guest on June 9, 2015psp.sagepub.comDownloaded from

940 Personality and Social Psychology Bulletin 41(7)
or their average cycle length. Likewise, given that we used
self-reported eating behaviors rather than objective eating
behaviors in Study 3, some participants may have inaccurately
perceived or reported the extent to which they restricted their
calories. In the absence of reason to expect such errors to be
systematic, however, any reporting errors should have reduced
our ability to detect effects, rather than contribute to them.
Second, our methods allow for the alternative conclusion that
women experience an increased motivation to consume more
calories during non-fertile periods of their ovulatory cycle
rather than a decreased motivation to consume fewer calories
near peak fertility. It is worth noting, however, that such an
interpretation seems unlikely given (a) the pattern of results
obtained in Studies 2 and 3, which revealed that naturally
cycling women differed from HC users at high rather than low
conception risk (see Figures 1 and 2) and (b) that conception
risk was unassociated with current weight in Studies 2 and 3.
Third, we utilized one-item measures to assess weight loss and
appearance motivations in Study 3. Although these measures
are high in face validity, and although the primary effect
obtained in that study conceptually replicated the effect that
emerged in Studies 1 and 2, future research may benefit by
examining the specific aspects of eating behavior that account
for the effects that emerged. Finally, although the current study
controlled for various factors (e.g., current weight, race, self-
esteem) and demonstrated a mechanism of the effect (e.g., the
motivation to manage body appearance), causal conclusions
should be drawn with caution.
Conclusion
The current research establishes that naturally cycling
women experience an increased motivation to lose weight
near peak fertility compared with less-fertile phases of their
ovulatory cycle and that this motivation is driven by wom-
en’s desires to manage the appearance of their bodies. Not
only do these findings indicate that the hormonal shifts that
occur across the ovulatory cycle can interact with modern
cultural norms regarding what is attractive, but they also sug-
gest such hormonal shifts can have implications for proximal
behaviors that on the surface may seem fairly far removed
from the ultimate goals they may function to serve (i.e.,
attaining and securing partners), such as the health-related
behaviors examined here.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support
for the research, authorship, and/or publication of this article:
Preparation for this article was supported by National Science
Foundation Grant BCS-1251520 to James K. McNulty.
Notes
1. The primary effect emerged as marginally significant when the
participant who was older than 35 was included, b = −0.39,
SE = 0.20, t(39) = 1.93, p = 0.061, 95% confidence interval (CI)
= [−0.79, 0.01], effect size r = .30.
2. In the interest of full disclosure, the primary analysis emerged
as non-significant when the participant who was older than 35
was included, b = 1.80, SE = 1.46, t(88) = 1.23, p = 0.220, 95%
CI = [−1.12, 4.72], effect size r = .13, but trended toward mar-
ginal significance when age was included as a covariate, b =
2.25, SE = 1.48, t(87) = 1.51, p = 0.134, 95% CI = [−0.71, 5.21],
effect size r = .16.
3. This broader sample was also used to test a different set of
hypotheses in another report (Baker & McNulty, 2015).
4. The primary effect remained significant when the six wives who
were older than 35 (M = 39.67, SD = 2.80) were included, b =
−0.16, SE = 0.07, t(92) = −2.26, p = 0.026, 95% CI = [−0.29,
−0.02], effect size r = .23.
5. One possible interpretation of this null effect, in light of the fact
that the predicted effect emerged as significant on the daily calo-
rie restriction variable, is that women who are highly motivated
to manage their body attractiveness and are not taking hormonal
contraceptives (HCs) are demonstrating an increased motivation
to manage weight (rather than to lose weight) at peak fertility.
We believe the results of Studies 1 and 2, where the primary
dependent variable was a net weight loss, suggest otherwise,
however. Thus, we suggest that, on a daily basis, women are
more aware of behaviors that can help them reach proximal
goals (i.e., reduced caloric intake) compared with behaviors
that can help them reach more ultimate goals (i.e., weight loss).
Indeed, most people do not lose weight in a day, but rather do
so over time by decreasing caloric intake (and increasing caloric
burn) on a daily basis. Thus, any motivations to lose weight that
were driven by ovulation in Study 3 may not have manifested
as daily desires to lose weight. Instead, such motivations likely
exist more ultimately, as assessed in Studies 1 and 2.
Supplemental Material
The online supplemental material is available at http://pspb.
sagepub.com/supplemental.
References
Andreoli, A., Monteleone, M., Van Loan, M., Proncenzio, L.,
Tarantino, U., & De Lorenzo, A. (2001). Effects of different
sports on bone density and muscle mass in highly trained ath-
letes. Medicine & Science in Sports & Exercise, 33, 507-511.
Baker, L. R., & McNulty, J. K. (2015). Adding insult to injury:
Partner depression moderates the association between part-
ner-regulation attempts and partners’ motivation to resolve
interpersonal problems. Personality and Social Psychology
Bulletin. Advance online publication.
Boniol, M., Autier, P., Boyle, P., & Gandini, S. (2012). Cutaneous
melanoma attributable to sunbed use: Systematic review and
meta-analysis. The British Medical Journal, 345, e4757.
Brown, S. G., Morrison, L. A., Calibuso, M. J., & Christiansen,
T. M. (2008). The menstrual cycle and sexual behavior:
Relationship to eating, exercise, sleep, and health patterns.
Women & Health, 48, 429-444.
by guest on June 9, 2015psp.sagepub.comDownloaded from

Meltzer et al. 941
Bryk, A. S., Raudenbush, S. W., & Congdon, R. T. (2004). HLM:
Hierarchical linear modeling with the HLM/2L and HLM/3L
programs. Chicago, IL: Scientific Software International.
Buss, D. M. (1988). From vigilance to violence: Tactics of
mate retention in American undergraduates. Ethology and
Sociobiology, 9, 291-317.
Calogero, R. M. (2004). A test of objectification theory: The effect
of the male gaze on appearance concerns in college women.
Psychology of Women Quarterly, 28, 16-21.
Cantú, S. M., Simpson, J. A., Griskevicius, V., Weisberg, Y. J.,
Durante, K. M., & Beal, D. J. (2014). Fertile and selectively
flirty: Women’s behavior toward men changes across the ovu-
latory cycle. Psychological Science, 25, 431-438.
Czaja, J. A., & Goy, R. W. (1975). Ovarian hormones and food
intake in female guinea pigs and rhesus monkeys. Hormones
and Behavior, 6, 329-349.
Durante, K. M., Griskevicius, V., Hill, S. E., Perilloux, C., & Li,
N. P. (2011). Ovulation, female competition, and product
choice: Hormonal influences on consumer behavior. Journal
of Consumer Research, 37, 921-934.
Durante, K. M., Li, N. P., & Haselton, M. G. (2008). Changes
in women’s choice of dress across the ovulatory cycle:
Naturalistic and laboratory task-based evidence. Personality
and Social Psychology Bulletin, 34, 1451-1460.
Durante, K. M., Rae, A., & Griskevicius, V. (2013). The fluctu-
ating female vote: Politics, religion, and the ovulatory cycle.
Psychological Science, 24, 1007-1016.
Eastwick, P. W., & Finkel, E. J. (2012). The evolutionary armi-
stice: Attachment bonds moderate the function of ovulatory
cycle adaptations. Personality and Social Psychology Bulletin,
38, 174-184.
Edler, C., Lipson, S. F., & Keel, P. K. (2007). Ovarian hormones
and binge eating in bulimia nervosa. Psychological Medicine,
37, 131-142.
Edwards, J. R. (1994). Regression analysis as an alternative to dif-
ference scores. Journal of Management, 20, 683-689.
Fallon, A. E., & Rozin, P. (1985). Sex differences in perceptions
of desirable body shape. Journal of Abnormal Psychology, 94,
102-105.
Fazio, R. H. (1990). Multiple processes by which attitudes
guide behavior: The mode model as an integrative frame-
work. Advances in Experimental Social Psychology, 23,
75-109.
Fehring, R. J., Schneider, M., & Raviele, K. (2006). Variability
in the phases of the menstrual cycle. Journal of Obstetric,
Gynecologic, & Neonatal Nursing, 35, 376-384.
Fessler, D. M. (2003). No time to eat: An adaptationist account of
periovulatory behavioral changes. The Quarterly Review of
Biology, 78, 3-21.
Fleischman, D. S., Navarrete, C. D., & Fessler, D. M. (2010).
Oral contraceptives suppress ovarian hormone production.
Psychological Science, 21, 750-752.
Frank, T. C., Kim, G. L., Krzemien, A., & Van Vugt, D. A. (2010).
Effect of menstrual cycle phase on corticolimbic brain activa-
tion by visual food cues. Brain Research, 1363, 81-92.
Funder, D. C., Levine, J. M., Mackie, D. M., Morf, C. C., Sansone,
C., Vazire, S., & West, S. G. (2014). Improving the depend-
ability of research in personality and social psychology:
Recommendations for research and educational practice.
Personality and Social Psychology Review, 18, 3-12.
Gangestad, S. W., & Thornhill, R. (2008). Human oestrus.
Proceedings of the Royal Society B: Biological Sciences, 275,
991-1000.
Gangestad, S. W., Thornhill, R., & Garver-Apgar, C. E. (2005).
Adaptations to ovulation: Implications for sexual and social
behavior. Current Directions in Psychological Science, 14,
312-316.
Garver-Apgar, C. E., Gangestad, S. W., & Thornhill, R. (2008).
Hormonal correlates of women’s mid-cycle preference for the
scent of symmetry. Evolution & Human Behavior, 29, 223-232.
Gildersleeve, K., Haselton, M. G., & Fales, M. R. (2014). Do wom-
en’s mate preferences change across the ovulatory cycle? A
meta-analytic review. Psychological Bulletin, 140, 1205-1259.
Gong, E. J., Garrel, D., & Calloway, D. H. (1989). Menstrual cycle
and voluntary food intake. The American Journal of Clinical
Nutrition, 49, 252-258.
Griffin, D., Murray, S., & Gonzalez, R. (1999). Difference score
correlations in relationship research: A conceptual primer.
Personal Relationships, 6, 505-518.
Harris, C. R., Chabot, A., & Mickes, L. (2013). Shifts in methodol-
ogy and theory in menstrual cycle research on attraction. Sex
Roles, 69, 525-535.
Haselton, M. G., & Gangestad, S. W. (2006). Conditional expres-
sion of women’s desires and men’s mate guarding across the
ovulatory cycle. Hormones and Behavior, 49, 509-518.
Haselton, M. G., Mortezaie, M., Pillsworth, E. G., Bleske-Rechek,
A., & Frederick, D. A. (2007). Ovulatory shifts in human
female ornamentation: Near ovulation, women dress to
impress. Hormones and Behavior, 51, 40-45.
Hebl, M. R., King, E. B., & Lin, J. (2004). The swimsuit becomes us
all: Ethnicity, gender, and vulnerability to self-objectification.
Personality and Social Psychology Bulletin, 30, 1322-1331.
Hill, S. E., & Durante, K. M. (2009). Do women feel worse to
look their best? Testing the relationship between self-esteem
and fertility status across the menstrual cycle. Personality and
Social Psychology Bulletin, 35, 1592-1601.
Hill, S. E., & Durante, K. M. (2011). Courtship, competition, and
the pursuit of attractiveness: Mating goals facilitate health-
related risk taking and strategic risk suppression in women.
Personality and Social Psychology Bulletin, 37, 383-394.
Hillhouse, J. J., Stair, A. W., & Adler, C. M. (1996). Predictors
of sunbathing and sunscreen use in college undergraduates.
Journal of Behavioral Medicine, 19, 543-561.
Jeffery, R. W., & Rick, A. M. (2002). Cross-sectional and longi-
tudinal associations between body mass index and marriage-
related factors. Obesity Research, 10, 809-815.
Kernan, W. N., Viscoli, C. M., Brass, L. M., Broderick, J. P., Brott,
T., Feldmann, E., & Horwitz, R. I. (2000). Phenylpropanolamine
and the risk of hemorrhagic stroke. New England Journal of
Medicine, 343, 1826-1832.
Lavine, H., Sweeney, D., & Wagner, S. H. (1999). Depicting
women as sex objects in television advertising: Effects on body
dissatisfaction. Personality and Social Psychology Bulletin,
25, 1049-1058.
Li, N. P., Smith, A. R., Griskevicius, V., Cason, M. J., & Bryan,
A. (2010). Intrasexual competition and eating restriction in
heterosexual and homosexual individuals. Evolution & Human
Behavior, 31, 365-372.
Li, N. P., Yong, J. C., Tov, W., Sng, O., Fletcher, G. J. O., Valentine,
K. A., . . . Balliet, D. (2013). Mate preferences do predict
by guest on June 9, 2015psp.sagepub.comDownloaded from

942 Personality and Social Psychology Bulletin 41(7)
attraction and choices in the early stages of mate selection.
Journal of Personality and Social Psychology, 105, 757-776.
Lord, F. M. (1967). A paradox in the interpretation of group com-
parisons. Psychological Bulletin, 68, 304-305.
McEachan, R. R. C., Conner, M., Taylor, N. J., & Lawton, R. J.
(2011). Prospective prediction of health-related behaviours
with the theory of planned behaviour: A meta-analysis. Health
Psychology Review, 5, 97-144.
Meltzer, A. L., & McNulty, J. K. (2015). Telling women that men
desire women with bodies larger than the thin-ideal improves
women’s body satisfaction. Social Psychological & Personality
Science, 6, 391-398.
Meltzer, A. L., McNulty, J. K., Jackson, G. L., & Karney, B. R.
(2014). Sex differences in the implications of partner physical
attractiveness for the trajectory of marital satisfaction. Journal
of Personality and Social Psychology, 106, 418-428.
Meltzer, A. L., Novak, S. A., McNulty, J. K., Butler, E. A., &
Karney, B. R. (2013). Marital satisfaction predicts weight gain
in early marriage. Health Psychology, 32, 824-827.
Miller, S. L., & Maner, J. K. (2010). Scent of a woman: Men’s tes-
tosterone responses to olfactory ovulation cues. Psychological
Science, 21, 276-283.
Mori, D., Chaiken, S., & Pliner, P. (1987). “Eating lightly” and
the self-presentation of femininity. Journal of Personality and
Social Psychology, 53, 693-702.
Morry, M. M., & Staska, S. L. (2001). Magazine exposure:
Internalization, self-objectification, eating attitudes, and body
satisfaction in male and female university students. Canadian
Journal of Behavioural Science, 33, 269-279.
Murayama, K., Pekrun, R., & Fiedler, K. (2014). Research practices
that can prevent an inflation of false-positive rates. Personality
and Social Psychology Review, 18, 107-118.
Must, A., Spadano, J., Coakley, E. H., Field, A. E., Colditz, G.,
& Dietz, W. H. (1999). The disease burden associated with
overweight and obesity. Journal of the American Medical
Association, 282, 1523-1529.
Pietromonaco, P. R., Uchino, B., & Dunkel Schetter, C. (2013).
Close relationship processes and health: Implications of attach-
ment theory for health and disease. Health Psychology, 32,
499-513.
Pittler, M. H., Schmidt, K., & Ernst, E. (2005). Adverse events of
herbal food supplements for body weight reduction: Systematic
review. Obesity Reviews, 6, 93-111.
Rosenberg, M. (1965). Society and the adolescent self-image.
Princeton, NJ: Princeton University Press.
Rothman, K. J., Wise, L. A., Sørensen, H. T., Riis, A. H., Mikkelsen,
E. M., & Hatch, E. E. (2013). Volitional determinants and age-
related decline in fecundability: A general population pro-
spective cohort study in Denmark. Fertility and Sterility, 99,
1958-1964.
Saad, G., & Stenstrom, E. (2012). Calories, beauty, and ovulation:
The effects of the menstrual cycle on food and appearance-
related consumption. Journal of Consumer Psychology, 22,
102-113.
Striegel-Moore, R. H., Silberstein, L. R., & Rodin, J. (1986).
Toward an understanding of risk factors for bulimia. American
Psychologist, 41, 246-263.
Tiggemann, M., & McGill, B. (2004). The role of social com-
parison in the effect of magazine advertisements on women’s
mood and body dissatisfaction. Journal of Social & Clinical
Psychology, 23, 23-44.
Tofighi, D., & MacKinnon, D. P. (2011). RMediation: An R pack-
age for mediation analysis confidence intervals. Behavior
Research Methods, 43, 692-700.
Wilcox, A. J., Dunson, D. B., Weinberg, C. R., Trussell, J., &
Baird, D. D. (2001). Likelihood of conception with a single act
of intercourse: Providing benchmark rates for assessment of
post-coital contraceptives. Contraception, 63, 211-215.
by guest on June 9, 2015psp.sagepub.comDownloaded from