Adjusting Signals of Sexual Interest in the Most Recent Naturally Occurring Opposite-Sex Encounter in Two Different Contexts

Article (PDF Available) · April 2019with 472 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
DOI: 10.1037/ebs0000162
Cite this publication
Abstract
Sexual signaling is subject to manipulation, and miscommunication may occur because of biased interpretations of signals, or because of strategical downplaying of sexual interest (playing hard-to-get). In this paper, we examined initial perceptions of cues from opposite sex partners along with participant reported own sexual attraction and signaled attraction in their most recent natural occurring potentially sexual opposite-sex encounter. Data on heterosexual Norwegian male and female students were collected in two largely different social contexts (during Regular Study Period, Spring 2015: N = 224 and during Freshmen Weeks, Early Fall 2015: N = 211). Results show no indication of women playing hard-to-get, or of strategically downplaying signals of sexual attraction. There was evidence of male sexual overperception in Study 1, but this effect was not replicated in Study 2 mainly due to increased levels of sexual attraction in single, freshmen women in that particular social context. For both sexes, reported levels of signaled attraction strongly reflected reports of own sexual attraction. Predictors for who ended up having sex after the encounter differed for women and men. For women, ending up having sex was predicted by the other’s short-term mate value, being freshman, and level of perceived sexual interest from the other after the encounter. For men, ending up having sex was predicted merely by their history of casual sex. It is concluded that women and men adjust their signals of sexual attraction upward or downward relative to their felt attraction to prompt further communication and to gain more information. http://dx.doi.org/10.1037/ebs0000162
Advertisement
Evolutionary Behavioral Sciences
Adjusting Signals of Sexual Interest in the Most Recent
Naturally Occurring Opposite-Sex Encounter in Two
Different Contexts
Mons Bendixen, Leif Edward Ottesen Kennair, Robert Biegler, and Martie G. Haselton
Online First Publication, April 4, 2019. http://dx.doi.org/10.1037/ebs0000162
CITATION
Bendixen, M., Kennair, L. E. O., Biegler, R., & Haselton, M. G. (2019, April 4). Adjusting Signals of
Sexual Interest in the Most Recent Naturally Occurring Opposite-Sex Encounter in Two Different
Contexts. Evolutionary Behavioral Sciences. Advance online publication.
http://dx.doi.org/10.1037/ebs0000162
Adjusting Signals of Sexual Interest in the Most Recent Naturally
Occurring Opposite-Sex Encounter in Two Different Contexts
Mons Bendixen,
Leif Edward Ottesen Kennair,
and Robert Biegler
Norwegian University of Science and Technology
Martie G. Haselton
University of California, Los Angeles
Sexual signaling is subject to manipulation, and miscommunication may occur because of
biased interpretations of signals, or because of strategical downplaying of sexual interest
(playing hard-to-get). In this article, we examined initial perceptions of cues from opposite
sex partners along with participant reported own sexual attraction and signaled attraction in
their most recent naturally occurring potentially sexual opposite sex encounter. Data on
heterosexual Norwegian male and female students were collected in two largely different
social contexts (during Regular Study Period, Spring 2015: N224; during Freshmen
Weeks, Early Fall 2015: N211). Results show no indication of women playing
hard-to-get, or of strategically downplaying signals of sexual attraction. There was evidence
of male sexual overperception in Study 1, but this effect was not replicated in Study 2
mainly due to increased levels of sexual attraction in single, freshmen women in that
particular social context. For both sexes, reported levels of signaled attraction strongly
reflected reports of own sexual attraction. Predictors for who ended up having sex after the
encounter differed for women and men. For women, ending up having sex was predicted
by the other’s short-term mate value, being freshman, and level of perceived sexual interest
from the other after the encounter. For men, ending up having sex was predicted merely by
their history of casual sex. It is concluded that women and men adjust their signals of sexual
attraction upward or downward relative to their felt attraction to prompt further commu-
nication and to gain more information.
Public Significance Statement
Meetings between potential romantic partners are fraught with ambiguity. This
study investigates how people perceive cues from a potential partner and how one
adjusts one’s own signals to the situation. Both sexes adjusted their signals: men
were in general moderately attracted but signaled less, women were in general not
attracted but signaled more. These findings are relevant for a greater understanding
of sexual misperception and the psychology of sexual harassment.
Keywords: error management theory, sex differences, sexual misperception, coyness,
communication
Mons Bendixen, Leif Edward Ottesen Kennair, and
Robert Biegler, Department of Psychology, Norwegian
University of Science and Technology; Martie G. Hasel-
ton, Department of Communication Studies, University of
California, Los Angeles.
We are grateful to our colleague Jonathan D. Kim for
proofreading and to Kyrre Svarva at the IT-section at
Norwegian University of Science and Technology (NTNU)
for formatting machine-readable questionnaires and for his
assistance scanning the data. Finally, we acknowledge the
assistance of students at the Program for Professional Stud-
ies in Psychology at NTNU.
Correspondence concerning this article should be ad-
dressed to Mons Bendixen, Department of Psychology,
Norwegian University of Science and Technology,
7491 Trondheim, Norway. E-mail: mons.bendixen
@ntnu.no
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Evolutionary Behavioral Sciences
© 2019 American Psychological Association 2019, Vol. 1, No. 999, 000
2330-2925/19/$12.00 http://dx.doi.org/10.1037/ebs0000162
1
When meeting someone in a potentially ro-
mantic or sexual situation, there is often a mea-
sure of uncertainty with regard to the other’s
sexual intentions. Various observable cues are
used for making inferences about the other per-
son’s sexual interest, but these cues are often
indirect or subtle, conveying considerable
amounts of ambiguity. Although such ambigu-
ity increases the risk of misperception, direct or
unambiguous communications of sexual intent
can also be detrimental to the mate value of the
person taking the initiative if he or she is either
rejected or earns a reputation of being sexually
indiscriminate (i.e., promiscuous). Conveying
unambiguous sexual interest might be efficient
in achieving a short-term relationship for
women, however, it is less efficient for men
(Bendixen & Kennair, 2015; Schmitt & Buss,
1996). Moreover, some ambiguity in courtship
communication can foster further communica-
tion between the parties, allowing for further
information to be gained about the prospective
mate’s commitment or sexual interest, which
may create a stronger foundation for evaluation
and later decision-making (Jonason & Li,
2013).
Both men and women misinterpret sexual
interest in dating situations, however, this oc-
curs in different ways. There is substantial em-
pirical evidence that men are more likely than
women to perceive sexual intentions from the
opposite sex’s signals (Haselton & Galperin,
2013; La France, Henningsen, Oates, & Shaw,
2009). Men’s tendency to overperceive is found
across diverse methodologies such as in face-
to-face single dyadic interactions in the labora-
tory (participant and observer ratings), videos
and pictures of dyadic interactions, written sce-
narios, vignettes and statements, speed-dating
dyadic interactions, naturalistic experiences,
and experiments (Haselton & Galperin, 2013;
La France et al., 2009). Importantly, this pattern
is equally strong in religious cultures as it is in
secular and sexually liberal cultures (Bendixen,
2014).
Error management theory (EMT; Haselton &
Buss, 2000; Haselton & Nettle, 2006 provides a
framework for interpreting these results. Judg-
ments and decisions are made under uncertainty
across a number of domains, such as navigating
the physical environment, cooperation with oth-
ers, and intuiting the intentions of a potential
partner. On the basis of signal detection theory
(Green & Swets, 1966; Swets, Dawes, & Mo-
nahan, 2000), EMT describes how natural se-
lection may have engineered psychological ad-
aptations for judgment under uncertainty. In
addition to making correct judgments (true pos-
itives and true negatives), two types of judg-
mental errors can be committed: A person may
adopt a belief that is in fact not true (false
positive) or fail to adopt a belief that is in fact
true (false negative). Within domains in which
the costs of errors have been asymmetrical over
deep evolutionary time, selection may favor de-
signs that make the less costly error of the two.
When the reproductive costs of missed sexual
opportunities were greater than the costs of pur-
suing uninterested mating partners, natural se-
lection, according to EMT, would produce the
adaptively biased systems that exist in the pres-
ent as they led to survival and reproductive
advantages for humans in the past. Pursuit
might cover a variety of acts spanning from
subtle sexual advances to simple capturing.
However, the specific acts were not explicitly
stated by the authors of EMT. One outcome of
these adaptive biases is the decision to actively
pursue sexual opportunities, even though this
increases overall error rate. Because ancestral
men, much more than women, could have in-
creased their reproductive success by increasing
their number of matings, a cognitive bias lead-
ing to greater beliefs that women are sexually
interested more often than they truly are (biased
beliefs), or to act as if women were sexually
interested, would produce more errors overall
relative to a design that maximized correct clas-
sification rates. This tendency could still maxi-
mize the overall expected value of decisions, as
it would have led to fewer high-cost errors at the
expense of making more low-cost errors. On
average, ancestral women experienced fewer
costs from missed mating opportunities; and,
hence, selection should not have favored the
same tendency (Haselton & Buss, 2000).
Signal detection theory, on which EMT is
based, does not explicitly distinguish belief
thresholds from action thresholds. Such a dis-
tinction between cognitive and behavioral bi-
ases has been suggested by McKay and Effer-
son (2010). If that distinction is introduced,
however, one may ask whether a man making
sexual advances acts on genuine, though often
inaccurate, biased beliefs about a woman’s true
sexual intent, or whether a man’s actions are
2 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
influenced by other factors, such as the percep-
tion of the costs of pursuing a woman versus
passing up a possible opportunity to mate. If a
man justifies his unwelcome advances on the
grounds that the woman really did welcome
them, is that a genuinely mistaken belief, or is
that the socially acceptable excuse of a man
who knows what costs he imposes on a woman,
but does not care? Galperin and Haselton (2013)
acknowledged that the core logic of EMT does
not depend on whether a change in payoff
causes a change in behavior through a change in
belief or a change in the assessment of out-
comes. The logic holds so long as psychological
adaptations that lead the organism to make more
errors overall (in belief, behavior, or both) also
reduce overall costs to that organism (by pur-
suing uninterested potential mates or missed
mating opportunities). Nevertheless, they main-
tain that other empirical evidence suggests that
the sexual overperception bias is rooted in bi-
ased beliefs. Still, being able to separate biases
that are cognitive from those that are merely
behavioral is important because this addresses
the issue of whether adaptive biases predicted
by EMT involve biases in cognition.
From EMT we hypothesize male sexual over-
perception bias (H1). We predict that men re-
port more initial sexual interest from their part-
ner (women) compared to the level of sexual
attraction reported by women. This bias is ex-
pected to be lower or reversed in women; that
is, female sexual underperception (Bendixen,
2014; Haselton & Buss, 2000; Haselton &
Galperin, 2013).
Recently, Perilloux and Kurzban (2015)
questioned whether men’s sexual overpercep-
tion truly reflects a cognitive bias. They claimed
that such a bias is unlikely because it would be
associated with the costs of erroneous represen-
tations that serve as inputs into other decision-
making systems, thereby distorting accuracy in
those other domains. They suggested that men’s
perceptions (cognitions) should be relatively ac-
curate (not perfectly accurate), whereas their
behavior should be biased: “pursuing even low-
probability/high-payoff opportunities” (Peril-
loux & Kurzban, 2015, p. 71). Applying the
Dating Behavior Scale (DBS; Haselton & Buss,
2000), which measures intentions for various
hypothetical behaviors, Perilloux and Kurzban
(2015) reproduced the original findings, sug-
gesting that men’s ratings of women’s sexual
intent when engaging in DBS behaviors ex-
ceeded female participants’ ratings when they
hypothetically engaged in the same behaviors.
Further, they asked men and women to rate
separately the likelihood of (1) what women
actually intend (i.e., what women actually want
when they engage in the DBS behaviors) and
(2) what women would report that they intend
(i.e., what they say they want). There was no
sex difference in reports of what women actu-
ally intended performing the behaviors in ques-
tion. However, men were more likely than
women to believe that women would say their
sexual intentions were lower than their actual
intentions were while performing DBS behav-
iors, and the “say” and “want” scores differed
significantly more for men than for women.
Based on the above findings, Perilloux and
Kurzban (2015) interpret these as indication of
men’s biased behavior, rather than biased be-
liefs, and find it unlikely that men have evolved
information processing mechanisms to perceive
sexual intent in women when there is none. If
one assumes that men’s perception of women’s
sexual intentions is accurate, women’s explicit
reports of their sexual intentions when hypo-
thetically engaging in DBS behaviors must be
due to some form of motivated underreporting
or understating.
Attempts to reproduce Perilloux & Kurzban’s
findings have shown that manipulation of ques-
tion order affects the responses (Murray, Mur-
phy, von Hippel, Trivers, & Haselton, 2017).
Women rated their sexual intentions lower
when these questions appeared before rather
than after questions about what women say or
want, indicative of a “purer-than-thou” effect
(Engeler & Raghubir, 2018). Still, the preced-
ings findings suggest that people tend to believe
more sexual intent in women than women claim
(“say”) and that women self-report less sexual
intentions compared with what they report other
women intend when performing the same hy-
pothetical behaviors. The former may reflect
stereotypical beliefs about women understating
their sexual intent, the latter may reflect purer-
than-thou beliefs, or possibly strategically play-
ing hard-to-get (i.e., coyness) if the preceding
findings using hypothetical behaviors have their
parallel in actual behavior. However, this has
not yet been subject to study.
From the concerns raised by Perilloux and
Kurzban (2015), we hypothesize an alternative
3ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
male bias that is rooted in behavior bias rather
than in biased beliefs (alternative male bias
behavior hypothesis, H2). If men report that
they pursue women (e.g., by sending signals of
sexual attraction) regardless of the perceived
cues to sexual interest of the opposite sex part-
ner, this would be indicative of biased behavior
(Perilloux & Kurzban, 2015). One way to pur-
sue would be to signal own sexual attraction.
Women are not expected to do this.
Compared with conveying indirect or ambig-
uous signals, direct and distinct signaling of
sexual interest to attract a mate may be a less
optimal strategy. Coyness as a sexual strategy
tactic for attracting mates was first identified by
Darwin (1871) and reflects in the animal king-
dom strategic reluctance to mate (McNamara,
Fromhage, Barta, & Houston, 2009; Wacht-
meister & Enquist, 1999). In humans, the as-
sumed function is to create an impression of
limited sexual availability in potential mates’
minds, and thus may be an effective tactic for
increasing the demand for the hard-to-get per-
son (Jonason & Li, 2013). Within the frame-
work of EMT (Haselton & Buss, 2000) the
function of playing hard-to-get would be to both
gain more information about potential mates
and to test their level of commitment. It follows
from Trivers’ (1972) parental investment theory
and Buss and Schmitt’s (1993) sexual strategies
theory that allocation of resources from both
parents increases survival in human offspring,
and that both men’s and women’s predominant
sexual strategy for achieving this would be
long-term committed relationships. Still, be-
cause the minimal obligatory offspring invest-
ment is markedly higher for women, there are
higher costs for women mating a man who
provides low investment. Hence, to increase
demand, women can limit their availability by
signaling less interest in short-term encounters
and by expanding courtship periods, thus both
avoiding matings with noninvesting men and
gathering more information about potential in-
vestment interest and ability. A side effect of
this information gathering might be tactics such
as playing hard-to-get. Although men can also
limit their availability, there are heavier costs
for men than for women through the loss of
potential mating opportunities. On the other
hand, women should not be attracted to men
who signal high availability as this may be a
signal of future defection (Jonason & Li, 2013).
In a series of studies Jonason and Li (2013)
assessed frequency of use and reasons for using
hard-to-get tactics. They found that women
used such tactics slightly more than men (d
0.17), but that the various reasons for using
hard-to-get tactics did not differ for women and
men. When presented with hypothetical pro-
spective opposite sex mate (descriptions and
pictures) for casual sex, men were significantly
more likely than women to prefer the mate who
often goes out with someone they just met (high
availability), while women’s preferences for ca-
sual sex partners was less associated with de-
gree of availability. However, for a romantic
long-term relationship, men were more likely
than women to prefer a mate with low levels of
availability. These findings correspond well to
what has been reported in studies of self-
promotion tactics, suggesting that signaling low
availability is considered ineffective for women
in short-term mating contexts (Bendixen &
Kennair, 2015; Schmitt & Buss, 1996).
Finally, some evidence of coyness was pro-
vided by Murray and colleagues (2017, supple-
mentary material) who asked a large sample of
women how frequently they had acted (1) more
interested in sex than they really were and (2)
less interested in sex than they really were when
dating someone (response alternatives were
from 1 (never)to7(always). They were also
asked how often they thought other women
acted less or more interested. The women re-
ported that, compared to how they really felt,
they more often acted less than more interested,
suggesting coyness (utilization of the hard-to-
get tactic). However, reports of acting less as
well as more interested were common. The par-
ticipants also believed other women use these
tactics and understate or underreport their sex-
ual intentions more often than they do them-
selves.
Thus, we hypothesize that the women may
play hard-to-get, that is, they strategically act
coy. We predict that women, but not men, rate
their signaled sexual attraction lower than their
own (felt) sexual attraction when reporting from
their own opposite sex encounters (Jonason &
Li, 2013; Murray et al., 2017). We term this the
Female Acting Hard-to-Get Hypothesis (H3).
An alternative to the female hard-to-get hy-
pothesis could be that signals of sexual attrac-
tion will be contingent upon one’s own level of
sexual interest (Murray et al., 2017). Because
4 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
information gained about the prospective
mate’s commitment or sexual interest may cre-
ate a stronger foundation for evaluation (Hasel-
ton & Buss, 2000; Jonason & Li, 2013), we
predict that those strongly attracted will cur-
tail their signals, whereas those at the low end
of sexual attraction will step up their signals
to prompt further communication. We term
this the adjustment in signaled attraction hy-
pothesis (H4). Both men and women are ex-
pected to do so.
The Current Studies
We examined misperceptions of sexual intent
in opposite sex encounters, and tactical signal-
ing of own sexual attraction using self-reports
from naturalistic dating contexts. In two sepa-
rate studies—the second carried out to replicate
the first, but in a context of abundant mating
opportunities (i.e., freshmen weeks)— under-
graduate students reported on their most recent
opposite sex encounter, including ratings and
qualitative descriptions of their perceptions of
cues to others=interest, ratings of how sexually
attracted the participants themselves felt, and
also ratings and qualitative descriptions of their
signaled attraction toward their encounter part-
ner. A distinction can be made between cues
and signals. Signals are cues emitted for the
purpose of communicating information. Those
cues that are not signals are uncontrolled leak-
age of information. We posit that naturalistic
approach of the most recent encounter permits
better insight into the psychology of mispercep-
tion and coyness (i.e., playing hard-to-get) than
either the use of judgment of sexual intentions
from a list of hypothetical behaviors women
may perform to signal their sexual intent
(Haselton & Buss, 2000; Perilloux & Kurzban,
2015) or recall of the relative frequency of
acting less or more sexually interested (Murray
et al., 2017).
Study 1
Method
Participants. Participants were undergrad-
uate students attending lectures in Social and
Natural sciences at a Norwegian University in
March/April 2015 (mid-term). On the basis of
their pattern of responses to the misperception
questions six cases were identified as outliers.
These were removed from the data along with
participants who did not indicate strong sexual
preference for opposite sex partners. The final
sample eligible for analysis of sexual misper-
ception consisted of 141 heterosexual women
and 83 heterosexual men aged between 20 and
29 years. The average age of the women and
men was 22.1 (SD 2.0) and 23.3 (SD 2.1),
respectively. Half (50%) of the students re-
ported ‘Being partnered’ as their current rela-
tionship status (54% women, 42% men).
Procedure. Research assistants recruited
participants during lecture breaks. Participant
instructions read as follows: “The purpose of
this study is to gain knowledge on the sexual
interplay between women and men and how we
interpret cues from persons of the opposite sex.
The questionnaire covers questions on personal
characteristics, your interpretation of situations
where you interacted with others, what you did,
your sexual experiences, attitudes, and fanta-
sies.” Questionnaires were handed out with a
prestamped return envelope and the participants
completed the survey at home or in a private
setting. Participation was fully voluntary and
anonymous. No incentives were given for par-
ticipation.
Measurements.
Sexual attraction, interpreting cues, and
displaying signals. After reporting on demo-
graphics, participants were presented with the
following: “Take a minute and reflect on the last
time you were at a gathering, at a party, or at a
disco interacting with a member of the opposite
sex that was not your intimate partner. We want
you to consider cues or signals that you picked
up during the conversation (interaction) and
how you responded to those cues. These cues
could indicate someone trying to be friendly
(just trying to be nice), showing sexual interest
(hitting on you) or something else.” Participants
rated the following: “Based on the cues s/he
sent me I initially assumed s/he was...using
a seven-point response scale with anchors 3
(just trying to be nice)and3(definitely sexually
interested) and mid-point 0 (didn’t know). We
have named this Perceived (Initial) Interest.
Then they were asked to write down a short
qualitative description of the cues they received.
These were coded into four qualitative distinct
predefined categories reflecting the explicitness
of the cues by two independent raters: The first
5ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
category, no sexual interest (coded 1) contained
“dismissive,” “informed about boy/girlfriend,”
“nice and kind,” “friendly,” “general conversa-
tion,” and so forth.
1
The second category, some
sexual interest (coded 2) contained “eye con-
tact,” “smile and eye contact,” and so forth).
The third category, moderate sexual interest
(coded 3) contained “smiling and body con-
tact,” “flirting,” “touching with compliments,”
and so forth, and the final category, strong sex-
ual interest (coded 4) contained “hard on,”
“pinched my bottom,” “touched my crotch,”
“wanted to go to somewhere private,” and so
forth The correlation with the seven-point re-
sponse scale, Perceived (Initial) Interest, was
substantial (r.64). Next, each participant
rated “How sexually attracted did you feel to-
ward the other person?” on a 7-point response
scale with the anchors 3(I did not feel sexu-
ally attracted)and3(I felt sexually attracted),
and midpoint 0 (I didn’t know). We have named
this Own Attraction.
Next the participants were asked the follow-
ing: “What did you signal?” They rated their
responses on a seven-point scale with the an-
chors 3(I was not sexually attracted), and3(I
was definitely sexually attracted). We have
named this signaled attraction. They were then
asked to write down a short qualitative descrip-
tion of the signals they conveyed. These were
coded into four qualitative categories of sig-
naled attraction, similar to the coding of cues
received.
2
The correlation with the seven-point
response scale signaled attraction was substan-
tial (r.67). Finally, we asked the participants:
“After you signaled, what did you learn about
his/her intentions?” A seven-point response
scale was used with the anchors 3(s/he was
just trying to be nice)and3(s/he was definitely
sexually interested) and mid-point 0 (I never
knew). This measure was named perceived in-
terest after.
Additional measures. Participants rated the
other party’s attractiveness (mate value) as po-
tential (1) casual sex partner: “How attractive
did you find the other person for short-term,
casual sex (a one-night stand)?” and (2) long-
term partner: “How attractive did you find the
other person as a long-term marriage partner.”
A seven-point response scale was used with
anchors 1 (well below average)and7(well
above average). Similarly, participants rated
their own level of attractiveness as a casual sex
partner and long-term marriage partner. For
measuring preference for casual, short-term
sexual relations, we applied the revised nine-
item Sociosexuality Orientation Inventory (SOI;
Penke & Asendorpf, 2008). The nine items re-
flect three dimensions or domains: behavior,
attitudes and desires/fantasies that each were
internally consistent: behavior (␣⫽.87), atti-
tudes (␣⫽.84), and desire (␣⫽.89). Scaling
and scoring followed the recommendations by
Penke and Asendorpf (2008).
All statistical analyses were performed using
Stata MP Version 15.1 for Mac (StataCorp,
2017). For interpreting effect sizes, we have
used Cohen’s (1988) conventions.
Results and Discussion
Indicators of sexual misperception. To
examine the male sexual overperception bias
hypothesis (H1) we compared the perceived
initial interest from the opposite sex party for
each sex with sexual attraction as reported by
the opposite sex. As can be seen from Table 1,
men’s perception of women’s interest was
higher than women’s self-reports of their own
sexual attraction, t(222) 6.11, p.001, d
0.84. In comparison, women’s perception of
men’s interest did not differ significantly from
men’s self-reported sexual attraction, t(220)
1.08. Although this is not a direct test, this
pattern of finding is indicative of sexual over-
perception in men.
These findings are supportive of H1; men
clearly overperceive the women’s interest,
whereas women’s perceptions are not biased.
We note that these are indirect tests, as these
reported naturally occurring encounters did not
permit studying dyads. The interpretation also
rests on the assumption that women and men
report accurately on their level of sexual attrac-
tion. Furthermore, we cannot rule out the pos-
sibility that some of the participants have re-
ported on encounters with someone outside the
1
Interrater reliability, kappa for perceived (initial) inter-
est was 0.65 (95% CI [0.59, 0.72]), suggesting substantial
agreement (McHugh, 2012). Agreement between raters was
approximately 80%. Disagreements were resolved by dis-
cussion.
2
Kappa for signaled attraction was 0.69 (95% CI [0.63,
0.75]), suggesting substantial agreement (McHugh, 2012).
Agreement between raters was approximately 80%. Dis-
agreements were resolved by discussion.
6 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
student population. Still, the test may be con-
sidered relevant as long as the samples of men
and women reflect their social context and that
the social context is equal and representative for
women and men.
Participant sex and relationship status ef-
fects on perceived initial interest, sexual at-
traction, signaled attraction, and perceived
interest after the encounter. As evident
from Table 1, the means for women and men’s
ratings of perceived initial interest from the
opposite sex did not differ much between sexes,
and for singles versus partnered participants. A
2 (Participant sex) 2 (Relationship status)
analysis of variance (ANOVA) revealed a non-
significant participant sex effect, F(1, 219)
2.68, and a marginally significant relationship
status effect, F(1, 219) 3.55, p.061, p
2
.016. Single participants rated the other per-
son’s initial interest slightly more ‘sexual’ than
partnered participants, but perceived opposite
sex interest was close to “didn’t know” (neutral)
across the four groups.
Regarding own sexual attraction men re-
ported on average on the midpoint of the scale,
whereas women on average reported their own
sexual attraction in the lower end of the scale.
This participant sex effect was significant and
large, F(1, 219) 44.59, p.001, p
2.169.
Compared with single participants, own sexual
attraction was markedly lower among those
partnered, F(1, 219) 56.42, p.001, p
2
.205. Furthermore, women signaled (sent) mod-
erately less attraction toward the other than
men, F(1, 220) 10.65, p.001, p
2.046,
and partnered participants (M⫽⫺2.03) sig-
naled markedly less attraction than singles
(M⫽⫺0.54), F(1, 220) 46.32, p.001,
p
2.174.
The perceived interest from the opposite sex
after the exchange of signals did not differ be-
tween women and men, F(1, 220) 1.19, but
singles reported the other party to be moderately
more interested than those partnered, F(1,
220) 19.56, p.001, p
2.082. None of the
analyses above revealed any Participant Sex
Relationship Status interaction effects (i.e., the
differences between women and men were not
moderated by relationship status or vice versa).
Signaled attraction as predicted by per-
ceived initial interest. To test the alternative
male biased behavioral hypothesis (H2), we re-
gressed signaled attraction on perceived initial
interest, participant sex and relationship status.
The three predictors accounted for 34.9% of the
variance in signaled attraction. Perceived initial
interest evinced a strong and positive associa-
tion with signaled attraction (␤⫽.37, t6.28,
p.001). When perceived initial interest was
accounted for, men signaled more attraction
than women (␤⫽.24, t4.25, p.001) and
singles more than those partnered (␤⫽⫺.35,
t⫽⫺6.21, p.001). The model (unstandard-
ized betas) suggests that men’s scores were on
average 0.88 (SE 0.21) higher than women’s,
and those partnered scored on average 1.23
(SE 0.20) lower than singles. The effect of
perceived initial interest was not moderated by
participant sex or relationship status, suggesting
that the associations between perceived initial
interest and signaled attraction were similar in
these subgroups (rs ranged from .39 to .47).
Additional analysis suggests that the effect of
perceived initial interest on signaled attraction
was curvilinear (added quadratic term, t⫽⫺2.
42, p.05). As shown in Figure 1, when initial
cues clearly indicated sexual interest, both
women and men levelled off their signaled at-
traction. However, when initial cues indicated
“just friendly” (negative scores) women and
men signaled low levels of attraction, albeit
men signaled more overall.
Table 1
Self-Reported Interest and Attraction Means and SDs for Women and Men, Singles and Partnered, Study 1
Variable
Women (n141) M(SD) Men (n83) M(SD)
Single Partnered All Single Partnered All
Perceived interest (I) .35 (1.91) .17 (2.04) .07 (1.99) .10 (1.93) .63 (2.14) .33 (2.02)
Own attraction 1.08 (1.86) 2.51 (1.13) 1.85 (1.66) .55 (1.61) 1.26 (1.60) .22 (1.83)
Signaled attraction .97 (1.78) 2.16 (1.20) 1.61 (1.60) .04 (1.76) 1.74 (1.56) .71 (1.89)
Perceived interest (A) .52 (1.90) .67 (2.06) .12 (2.07) .25 (1.90) 1.00 (2.04) .28 (2.04)
Note.Iinitial; A after.
7ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Because there was no evidence of men sig-
naling high levels of sexual attraction regardless
of perceived level of interest from the woman,
Hypothesis 2 was not supported. Evidently,
both men’s and women’s signals of sexual in-
terest were strongly contingent on their initial
perception of interest from the opposite sex
party. Those encountering sexually uninterested
persons (initially perceived as friendly) signaled
low attraction, while those meeting up with
persons perceived as having some or strong
sexual interest sent moderate signals of sexual
attraction back (men more clearly than women).
Signaled attraction as predicted by own
sexual attraction. To test the female acting
hard-to-get hypothesis (H3) and the adjusting
signals of sexual attraction hypothesis (H4), we
first regressed signaled attraction on level of
sexual attraction, participant sex and relation-
ship status. Then we regressed in Model 1 the
difference in signaled attraction versus own sex-
ual attraction (i.e., adjustment of signals) on
participant sex and relationship status and ac-
counting for own sexual attraction in Model 2.
Positive scores reflect more sexual attraction
relative to signaled attraction (i.e., curtailing
one’s sexual attraction). In the first regression,
sexual attraction was a particularly strong pre-
dictor of signaled attraction (␤⫽.79, t14.71,
p.001), men appeared to signal less attraction
than women (␤⫽⫺.09, t⫽⫺2.01, p.05)
when own sexual attraction was accounted for,
while relationship status had no effect (t
1.44). The effect of sexual attraction on sig-
naled attraction was not moderated by partici-
pant sex or relationship status, suggesting that
the association between the attraction one felt
and level of signaled attraction was similar
across the four subgroups (rs ranging from .61
to .81).
In the second regression (adjustment of sig-
nals), men overall reported more sexual attrac-
tion than signaled attraction than did women in
Model 1 (␤⫽.29, t4.35, p.001). Post hoc
paired-samples ttests for women showed sig-
nificantly elevated levels of signaled attraction
relative to own sexual attraction, t(140) ⫽⫺2.
62, p.01. Men on the other hand, signifi-
cantly curtailed their level of signaled attraction
relative to own sexual attraction, t(81) 3.62,
p.01. Singles and partnered participants did
not differ in their overall level of adjustment.
When own sexual attraction was accounted for
in Model 2, men still adjusted (curtailed) their
signals more than women (␤⫽.14, t2.01,
p.05). As we can see from Figure 2, the more
own sexual attraction women and men felt, the
more they curtailed their signals (positive dif-
ference scores), and level of curtailing was
marked for both sexes at high levels of sexual
attraction. The association was moderate and
similarly strong for women (␤⫽.41) and men
(␤⫽.37), but at every level of own sexual
Figure 1. Level of signaled sexual attraction as a function of level of perceived initial
interest. Scatter plots and predicted fitted values [95% CIs]. Panel A: Women (n141);
Panel B: Men (n83).
8 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
attraction, men appeared to curtail their signals
more.
On average, women’s signaled attraction was
higher than their sexual attraction; it was men
who curtailed their sexual attraction. There was
no support for H3 that women play hard-to-get.
However, degree of curtailing followed predict-
able patterns for both men and women support-
ing H4. With increasing level of sexual attrac-
tion, men and women increasingly curtailed
their signals, with men doing this more at every
level of sexual attraction than women. At low
levels of own sexual attraction, women elevated
their signals, and signaled more than they felt.
For men, this makes sense based on Jonason
and Li (2013) and research on the perceived
efficiency of different self-promotion tactics
(Bendixen & Kennair, 2015; Schmitt & Buss,
1996)—it might be counterproductive for men
to actually signal their true sexual attraction,
particularly if their level of attraction is high; as
it might not be considered attractive. Women on
the other hand— by not sending signals of total
lack of sexual attraction—may keep the man’s
attention for longer, and maybe thereby get to
assess him further or strategically increase his
hope of having a chance (Jonason & Li, 2013).
This may be understood as a form of tactical
misrepresentation of own sexual attraction that
seems designed to effectively nurture further
interaction with the opposite sex.
Study 2
Study 2 was conducted to see whether or not
the general pattern of results from Study 1
would replicate in a context that increased the
likelihood of sexual encounters and abundant
mating opportunities. Specifically, Study 2 was
carried out during the university’s introduction
weeks of the academic year (end of August), a
period characterized by strong socialization and
partying (organized freshmen rituals and men-
tor group affiliation). The questions posed were
similar to those of Study 1, but we included
information on freshman status (first year stu-
dent) and posed additional questions on (1) who
initiated the contact and (2) the outcome of the
encounter. Study 2 sets out to test the same
hypotheses as in Study 1.
Given the apparent differences in opportuni-
ties for mating between Study 1 and Study 2, we
specifically wanted to examine to what extent
this affected aspects of participant’s sexual psy-
chology such as level of sexual attraction, sig-
naled attraction, judgments of partner mate
value and orientation toward short-term sexual
relations. In addition, we wanted to examine
factors that predict who ends up having sex in
the most recent opposite sex encounter, and to
what extent these factors were different for
women and men. To achieve this, we studied
how well singles’ self-reported sexual attraction
Figure 2. Adjustment of signals (level of own attraction—level of signaled attraction) as a
function of own sexual attraction. Negative adjustment scores: Signals Attraction; positive
adjustment scores: Signals Attraction. Scatter plots and predicted fitted values [95% CIs].
Panel A: Women (n141); Panel B: Men (n83).
9ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
and signaled interest predicted who ended up
having sex following the encounter, and to what
extent how much sexual interest (before and
after) they perceived from the other party af-
fected the probability of ending up having sex.
We also examined to what extent ending up
having sex was predicted by the respondent’s
preference for short-term sexual relations as
well as with the perceived mate value of the
partner. In general, we would expect own sexual
(and signaled) attraction along with cues to sex-
ual interest from the other part to increase the
probability ending up having sex in both men
and women. However, because men are less
discriminating and have lower minimum stan-
dards than women (Buss & Schmitt, 1993;
Regan, 1998) sex differences are expected in
what factors predict who ends up having sex
following the encounter. Sex differences in
minimum standards for casual sex would result
in men being relatively more often chosen and
women being the choosing party. We predict
that (1) perceptions of opposite sex partner’s
short-term attractiveness increases women’s
likelihood of having sex and (2) men with at-
tractive features for short-term mating have in-
creased likelihood of having sex.
Method
Participants, procedure, and measure-
ments. Participants were undergraduate stu-
dents attending lectures in Social and Natural
sciences at a Norwegian University in August/
September 2015. Three cases were identified as
outliers based on their pattern of responses to
the misperception questions. These were re-
moved from the data along with participants
who did not indicate strong sexual preference
for opposite sex partners. The final sample eli-
gible for analysis of sexual misperception con-
sisted of 135 heterosexual women and 76 het-
erosexual men aged between 19 and 30 years.
The average age of the women and men were
21.6 (SD 2.2) and 22.2 (SD 2.4) respec-
tively. Less than half (45%) of the students
reported ‘Being partnered’ as their relationship
status (47% women, 40% men) at the time of
the encounter. Half (51% of women and men)
of the students reported their academic status as
“freshman” (i.e., being enrolled this semester).
The procedure was identical to the one applied
to Study 1. Measurements were identical to
Study 1 with two additions: Before any ques-
tions regarding cues and signals, we asked who
initiated the encounter (“Who took contact
first”). Response alternatives were 1 (I did) and
2(the other person). Following these questions,
we asked about the outcome of the encounter
(no further contact/became acquainted/became
friends/kissed/had sex/became boyfriend or
girlfriend). Multiple responses were optional.
Among singles, 31% percent women (22 of 70)
and 20% men (9 out of 45) reported they “had
sex” following the most recent encounter.
Results and Discussion
Indicators of sexual misperception. To
test the male sexual overperception hypothesis
(H1) we compared, similar to Study 1, men’s
perceived initial interest from the opposite sex
party with women’s self-reported Sexual attrac-
tion and vice versa (see Table 2). Men’s per-
ception of women’s initial interest was not very
different from women’s self-reported sexual at-
traction, t(209) 1.11, indicating no sexual
overperception bias in men. Similarly, women’s
perception of men’s interest were only slightly
lower than men’s self-reported sexual attrac-
Table 2
Self–Reported Interest and Attraction Means and Standard Deviations for Women and Men, Singles and
Partnered, Study 2
Variable
Women (n135) M(SD) Men (n76) M(SD)
Single Partnered All Single Partnered All
Perceived interest (I) .08 (1.84) .03 (1.79) .03 (1.81) .53 (1.77) .47 (1.94) .50 (1.81)
Own attraction .34 (1.93) 2.11 (1.51) .82 (2.13) .84 (1.57) .53 (1.91) .32 (1.83)
Signaled attraction .18 (1.81) 1.80 (1.36) .76 (1.89) .07 (1.55) .90 (1.84) .32 (1.71)
Perceived interest (A) .73 (1.76) .06 (2.09) .41 (1.94) .29 (1.67) .67 (1.79) .41 (1.73)
Note.Iinitial; A after.
10 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tion, t(209) ⫽⫺1.10, indicating no overall un-
derperception bias in women. When we reran
the above analyses omitting freshmen students
(n107), men’s perceived initial interest (M
0.24) was significantly higher than self-
reported attraction by women (M⫽⫺1.09),
t(102) 2.12, p.05, d0.43). On the other
hand, women’s perceived initial interest (M
0.27) was significantly lower than the self-
reported sexual attraction by men (M0.76),
t(102) 2.72, p.01, d⫽⫺0.56).
Unlike Study 1, H1 was not supported in
Study 2. Neither men’s nor women’s perception
of the opposite sex showed bias; rather, they
corresponded closely to the opposite sex’ re-
ported sexual attraction. Further, sex differ-
ences in sexual attraction and signaled attraction
were smaller or nonsignificant in Study 2. As
the additional analyses of nonfreshmen students
in Study 2 and the above comparison analyses
suggest, the lack of support for H1 may be
attributable to contextual differences between
Study 1 and Study 2 with regard to the relative
presence of women and men, and of freshmen
and nonfreshmen. Women, but not men, found
the other party significantly more attractive as a
mate in Study 2 compared with Study 1. This
accounted for some of the study differences in
levels of own and signaled sexual attraction.
Apparently, the social contexts in which the
studies were carried out affected single wom-
en’s own and signaled attraction, while men’s
perceived opposite sex interest was unaffected
by the social context. If cost asymmetry was the
only factor, men should have adjusted, and con-
tinued to perceive more sexual interest than
women signaled. The first study was carried out
during a study-intensive period (the weeks be-
fore Easter), when the students have become
familiar with the university and the city. We
would denote this as the “normalized” state or
“baseline.” The social context of Study 2 was
characterized by a high proportion of freshmen,
including students who recently had moved
away from home for their studies, and who were
experiencing high levels of socializing through
arranged parties and gatherings (i.e., freshmen
rituals). Unmeasured aspects of differences in
social context related to partying, making new
friendships and relationships, and lax daughter
guarding might have affected the female sexual
psychology more than the male sexual psychol-
ogy (Perilloux, Fleischman, & Buss, 2008).
These findings might also reflect a relative
change in experienced/subjective operational
sex ratio between the two studies, possibly ac-
tivated by a high relative proportion of male
mentors for female-dominated groups that may
have provided sufficient signals to female sex-
ual competition to affect mate preferences to-
ward less choosiness in Study 2 (Hahn, Fisher,
DeBruine, & Jones, 2014; Kandrik, Jones, &
DeBruine, 2015; Moss & Maner, 2016; Okami
& Shackelford, 2001).
Participant sex and relationship status ef-
fects on perceived initial interest, own sexual
attraction, signaled attraction and perceived
interest after the encounter. A 2 (participant
sex: women 0 vs. men 1) 2 (relationship
status: singles 0 vs. partnered 1) ANOVA
on perceived initial interest revealed a signifi-
cant participant sex effect, F(1, 206) 3.91,
p.05, p
2.019, suggesting that women
perceived slightly more initial interest from the
opposite sex than did men. However, partnered
participants did not perceive sexual interest dif-
ferently from single participants, F(1, 206)
0.01. As seen from Table 2, on average men
reported their own sexual attraction close to the
midpoint of the scale and significantly higher
than women reported their own, F(1, 206)
16.97, p.001, p
2.076. Compared with
singles (M0.53), partnered participants (M
1.61) reported markedly less sexual attrac-
tion, F(1, 206) 57.25, p.001, p
2.217.
The effect of relationship status was qualified
by sex, F(1, 206) 4.48, p.05, p
2.021,
suggesting that the participant sex difference in
sexual attraction was stronger for partnered than
for single participants.
Counter to our general expectation, men did
not report significantly more signaled attraction
than women, F(1, 206) 2.67, p.104.
3
However, those who had a partner (M⫽⫺1.51)
signaled significantly less sexual attraction than
did those who were single (M0.14), F(1,
3
Subsequent analyses showed that the lack of sex differ-
ence was attributable to the influential effect of freshman
status on signaled attraction for women and men. Omitting
freshmen students from the analysis revealed a similarly
strong sex difference as those reported in Study 1,
F(1,99) 8.00, p.01, p
2.075 (women: M⫽⫺1.13;
men: M⫽⫺0.08). Further, the sex difference in sexual
attraction was clearly strengthened when omitting freshmen
from the analyses, F(1,99) 19.64, p.001, p
2.166
(women: M⫽⫺1.09; men: M0.76).
11ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
206) 38.11, p.001, p
2.155. The effect
of relationship status was qualified by sex, F(1,
206) 4.51, p.05, p
2.021, suggesting
that signaled attraction differed more for single
versus partnered women than for single versus
partnered men.
Finally, our analysis of perceived interest
from the opposite sex after the exchange of
signals showed that women perceived more op-
posite sex interest than men, F(1, 206) 10.46,
p.001, p
2.048. Relative to single partic-
ipants, perceived interest after the encounter
was marginally lower among those partnered,
F(1, 206) 3.75, p.054, p
2.018.
Signaled attraction as predicted by per-
ceived initial interest. As for Study 1, we
tested the alternative male biased behavior hy-
pothesis (H2) and regressed signaled attraction
on level of perceived initial interest, participant
sex, and relationship status. All predictors pro-
duced significant effects, accounting for 30.4%
of the variance in signaled attraction. Perceived
initial interest evinced a moderately strong and
positive association with signaled attraction
(␤⫽.32, t4.69, p.001). Over and above
the effect of perceived initial interest, men sig-
naled more than women (␤⫽.13, t2.17, p
.05) and those partnered signaled less than sin-
gles (␤⫽⫺.44, t⫽⫺7.40, p.001). The
model suggests that men scored on average 0.49
(SE 0.22) higher than women, and those
partnered scored on average 1.61 (SE 0.22)
lower than singles (unstandardized betas). The
effect of perceived initial interest was not moder-
ated by sex or relationship status, suggesting that
the associations between perceived interest and
signaled attraction were similarly strong across
these subgroups. Similar to Study 1, additional
analysis suggests the effect of initial perceived
interest was curvilinear (t⫽⫺2.45, p.05) and
that both women and men levelled off their sig-
naled attraction at high levels of Initial perceived
interest (as illustrated in Figure 1).
The above findings replicate those of Study 1,
and Hypothesis 2 (male biased behavior) was
neither supported in Study 2. Respondents’ sig-
naled attraction were clearly contingent on their
perception of initial signaled interest of the
other party. Specifically, men meeting uninter-
ested women signaled very little attraction.
Signaled attraction as predicted by own
sexual attraction. We reran the relevant tests
from Study 1 for the female acting hard-to-get
hypothesis (H3) and the adjusting signals of
sexual attraction hypothesis (H4). In the first
regression, sexual attraction again was the prin-
cipal predictor of signaled attraction (␤⫽.82,
t17.71, p.001). When accounting for the
effect of sexual attraction, men appeared to sig-
nal less attraction than women did (␤⫽⫺.10,
t⫽⫺2.39, p.05), whereas relationship status
had no effect (t⫽⫺0.64). Similar to Study 1,
the association between sexual attraction and
signaled attraction was not moderated by sex or
relationship status.
In the second regression (difference scores),
relative to women, men on average reported
more sexual attraction than signaled attraction
(␤⫽.25, t3.62, p.001). In addition,
singles reported less attraction than partnered
participants did (␤⫽⫺.18, t⫽⫺2.67, p
.01). The effect of participant sex was not mod-
erated by relationship status (and vice versa).
Post hoc paired-samples ttest for women re-
ported levels of signaled attraction no different
from sexual attraction, t(134) ⫽⫺0.70. Men on
the other hand, significantly curtailed their level
of signaled attraction relative to their own sex-
ual attraction, t(75) 4.22, p.001, and, at
every level of own sexual attraction, men sup-
pressed their signals more than women. The
association between level of own sexual attrac-
tion and the degree of curtailing was linear,
moderate, and similarly strong for women (␤⫽
.51) and men (␤⫽.44). Hence, both men and
women downplayed their signals at high levels
of sexual attraction. These findings paralleled
those of Study 1 (as illustrated in Figure 2).
The preceding findings do not support the
female acting hard-to-get hypothesis (H3). As
in Study 1, women’s own sexual attraction did
not show any overall difference from their sig-
naled attraction. However, the adjusting signals
hypothesis (H4) was again strongly supported.
Women signaled more attraction at low levels
of own attraction, while both men and women
curtailed their signals at high levels of sexual
attraction. Again, this adjustment of signals
seems designed to effectively nurture further
interaction with the opposite sex (Jonason & Li,
2013).
Predicting who ended up having sex in
Study 2. We first examined all relevant pre-
dictors for who ended up having sex for single
participants. These predictors included fresh-
man status, who took initiative in establishing
12 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
contact, perceived interest before and after, sex-
ual and signaled attraction, own short-term mate
value, the other party’s short-term and long-
term mate value, and each of the three socio-
sexuality dimensions (behavior, attitudes, de-
sire). As can be seen from Table 3, freshmen
women, women who met attractive men for
short-term relationships, women who perceived
high interest from the man after the encounter,
and women who were sexually attracted to the
man (and signaled accordingly) were more
likely to report ending up having sex. For single
men, having a history of many short-term sex-
ual partners (SOI–Behavior) and perceived sex-
ual interest from the woman after the encounter
were both associated with higher likelihood of
reporting ending up having sex. The effect of
being a freshman and the opposite sex partner’s
short-term mate value on ending up having sex
differed significantly between women and men.
Hence, separate logistic regressions models for
women and men were applied for predicting
ending up having sex.
For single women, the most parsimonious
model included three predictors: The other’s
short-term mate value (z2.25, p.05), being
a freshman (z1.96, p.05), and level of
perceived sexual interest from the other after the
initial encounter (z2.79, p.001) all in-
creased the probability of reporting ending up
having sex. Relative to nonfreshmen women,
freshmen women’s odds of ending up having
sex was 3.7 times higher. Notably, freshman
status had some moderating effect on the effect
of the other’s short-term mate value on likeli-
hood of ending up having sex (z⫽⫺1.84, p
.065). As illustrated in Figure 3, the other’s
mate value appeared to affect the probability of
ending up having sex more strongly for fresh-
men than for nonfreshmen single women.
Although single women’s own sexual attrac-
tion, their signals of attraction, SOI-behavior
and SOI-attitudes were all significantly associ-
ated with the outcome, none of these variables
had any effect over and above the variables in
the above model. The predictors accounted for
substantial variance (McFadden R
2
0.469),
and the model correctly classified 81.2% of the
cases with an equal number of false positives
and negatives. For single men, only SOI–
Behavior had any influence on the outcome (z
2.63, p.01) and the model correctly classified
77.8% of the cases with a higher rate of false
negatives (underidentifying those who ended up
having sex). The variance accounted in this
model for was about half that of the model of
women (McFadden R
2
0.216).
The above findings are in line with our pre-
dictions that perceptions of opposite sex part-
ner’s short-term attractiveness increases wom-
en’s likelihood of having sex, and that men with
attractive features for short-term mating have
increased likelihood of having sex. Single
women who ended up having sex reported hav-
Table 3
Bivariate Associations (Pearson’s R) and Fisher’s Z for the Sex Difference in
Correlations for Predictors of Ending Up Having Sex (No/Yes),
Single Participants
Predictor Women (n69) Men (n45) Fisher’s z
Freshman (0 no, 1 yes) .31
ⴱⴱ
.09 2.08
Initiative (1 self, 2 other) .11 .05 .31
Perceived interest (I) .20 .22 .11
Own attraction .44
ⴱⴱ
.27 .99
Signaled attraction .43
ⴱⴱ
.23 1.14
Perceived interest (A) .48
ⴱⴱ
.32
.97
Own male value (ST) .28
.22 .32
The other’s mate value (ST) .47
ⴱⴱ
.12 3.20
ⴱⴱ
The other’s mate value (LT) .06 .01 .10
SOI–Behavior .27
.46
ⴱⴱ
1.12
SOI–Attitudes .34
ⴱⴱ
.33
.06
SOI–Desire .23 .01 1.14
Note.Iinitial; A after; ST short-term; LT long-term; SOI Sociosexuality
Orientation Inventory.
p.05.
ⴱⴱ
p.01.
13ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ing met a man that they perceived as more
attractive for short-term sexual relations. In ad-
dition, they perceived the signals from him in-
dicating a clear sexual interest after the initial
encounter. Finally, these women were more
likely to be freshmen. None of these factors had
a significant effect on men’s likelihood of end-
ing up having sex. This dovetails neatly with
recent findings from the casual sex regret liter-
ature, where there was a particularly strong sex
differentiated effect of taking the initiative on
casual sex action regret (Kennair, Wyckoff,
Asao, Buss, & Bendixen, 2018). They found
less casual sex regret among women who took
the initiative. These was no such effect for men.
Probably, in light of the current results, less
regret when women take initiative may be due
to relatively higher mate value of the men with
whom they decided to have sex. The only rele-
vant factor for men was their unrestricted sexual
behavior; their history of casual sex behavior, or
their number of past partners. These findings
strongly support the sexual strategy theory’s
predictions (Buss & Schmitt, 1993; Regan,
1998), suggesting that women, being the choos-
ier sex, act upon their partner preferences and
mate with men with characteristics matching
these preferences.
Additional comparisons across studies:
Partner’s mate value, sociosexuality, sexual
attraction, and signaled attraction. Because
the two studies were carried out in contexts that
differed largely with respect to mating opportu-
nities—Study 1 during the spring term, when
most students prepare for their exams, and
Study 2 shortly after the ‘Freshmen’ weeks in
the beginning of the fall semester—we wanted
to examine the extent to which our findings
were affected by the increased opportunities for
mating in Study 2. In particular, we wanted to
examine (1) the extent to which own mate value
(short-term and long-term), the mate value of
the other party (short-term and long-term) as
reported by the participant, and participants’
sociosexuality (SOI–Behavior, SOI–Attitudes,
and SOI–Desire) differed across studies and (2)
how this may have affected observed sex dif-
ferences in Sexual attraction and Signaled at-
traction between Study 1 and Study 2.
To examine study differences, we performed
three-way factorial ANOVAs with Mate value
(own and other’s short-term and long-term) and
each dimension of sociosexuality as dependent
variables and study, participant sex, and rela-
tionship status as predictors. (The means and
standard deviations for the seven outcomes
across the groups are presented in Appendix A
and Appendix B.) Across the seven compari-
sons, significant study differences were found
for the other’s short-term mate value and SOI–
Attitudes only, suggesting that relative to Study
1, participants in Study 2 found their opposite
sex party more attractive for casual sex, F(1,
424) 6.43, p.05, and their attitudes toward
casual sex were less restricted, F(1, 424)
14.73, p.001). The study difference in the
opposite sex party’s short-term mate value was
moderated by participant sex, F(1, 423) 8.69,
p.01. Compared with women in Study 1
(M3.25), women in Study 2 (M4.04)
found the man markedly more attractive for
casual sex, whereas men’s scores of the wom-
an’s short-term mate value did not differ across
studies (M4.89 and M4.72 for Study 1
and Study 2, respectively). Notably, the other
party’s long-term mate value, own short-term
and long-term mate value did not differ between
studies, nor did SOI–Behavior or SOI–Desire.
However, some interactions did involve study:
The participant sex effect for the other’s long-
term mate value was moderated by study, F(1,
427) 5.77, p.05. In Study 1, men (M
4.14) found their partner significantly more at-
tractive as a marriage partner than women (M
3.13). In Study 2, this sex difference was mark-
edly reduced (Means were 3.83 and 3.56 for
men and women, respectively). Hence, whereas
Figure 3. Estimated probability of ending up having sex
for freshmen (n40) and nonfreshmen (n30) single
women as a function of perceived short-term mate value of
the opposite sex partner.
14 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
women in Study 2 appeared to find their partner
more attractive with regard to both short-term
and long-term relationships, men found their
partner slightly less attractive in Study 2.
4
To examine how strongly study differences
in the other party’s mate value affected partic-
ipants’ sexual attraction, signaled attraction,
and observed differences across studies, we re-
gressed sexual attraction and signaled attraction
separately on study, participant sex and rela-
tionship status (Model 1), adding partner’s
short-term and long-term mate value in Model
2. Interactions and nonlinear effects are re-
ported throughout. In Model 1, sexual attraction
was higher in Study 2 (␤⫽.18, t4.68, p
.001), men were more attracted than women
(␤⫽.29, t7.18, p.001), and partnered
participants were less attracted than singles
(␤⫽⫺.45, t⫽⫺11.27, p.001). Single
women in Study 2 were markedly more sexually
attracted than any of the other three groups of
women, producing a significant three-way
Study Participant Sex Relationship Status
interaction effect (␤⫽.18, t2.13, p.05).
In Model 2, both short-term (␤⫽.35, t7.11,
p.001) and long-term mate value of the other
party (␤⫽.17, t3.88, p.001) predicted
sexual attraction over and above the effect of
the Model 1 predictors. Adding mate value re-
duced the effect of the Model 1 predictors to
some extent (study: ␤⫽.13, t3.71, p.001;
participant sex: ␤⫽.15, t3.76, p.001; and
relationship status: ␤⫽⫺.34, t⫽⫺9.21, p
.001). The effects of the other party’s short-term
and long-term mate values were both moderated
by participant sex (p.10). Post hoc compar-
isons of correlations for women and men sug-
gest that sexual attraction was more strongly
correlated with the other party’s short-term
(r
women
.60, r
men
.33, Fisher’s z3.46,
p.001) and long-term mate value (r
women
.47, r
men
.22, Fisher’s z2.91, p.01) for
women. In addition, both measures of mate
value predicted sexual attraction in a curvilinear
way (significant quadratic terms).
5
The effect of
the other party’s short-term mate value on sex-
ual attraction is illustrated in Figure 4 separately
for Study 1 and Study 2 (Similar effects were
found for long-term mate value, albeit less
strong). Evidently, in both studies, women re-
ported being less sexually attracted than men
when meeting a low-to-moderately attractive
person of the opposite sex, but both men and
women reported equally strong sexual attraction
when meeting a highly attractive partner.
Subsequent analyses were performed on sig-
naled attraction. These reproduced closely the
findings of own sexual attraction above with
some notable exceptions. First, in Model 2,
once the effect of the other person’s mate value
(short-term and long-term) were accounted for,
men no longer signaled more attraction than
women (this difference was significant, ␤⫽
.14, t3.19, p.01, in Model 1, with men
scoring 0.53 units above women). Second, par-
ticipant sex did not moderate the effect of part-
ner’s short-term mate value on signaled attrac-
tion. Finally, when meeting a very attractive
person, women appeared to signal equal or
more sexual attraction than men. Still, when
meeting a low-to-moderately attractive partner,
women signaled markedly less than men.
General Discussion
In two separate studies, we examined misper-
ceptions of sexual intent in opposite sex en-
counters and tactical signaling of own sexual
attraction using self-reports from two different
naturalistic dating contexts by asking men and
women to report on their most recent opposite
sex encounter. In particular, we searched for
evidence that women play hard-to-get or act coy
by signaling less sexual attraction than they feel
in opposite sex encounters and that men may
pursue women regardless of cues to sexual in-
terest. If women generally act coy this would
possibly account for the phenomenon known as
male sexual overperception bias (Haselton &
Buss, 2000). We did not find any indication of
general female coyness, or that women curtail
or underreport their sexual attraction in either
study. Instead, we found that both men and
women adjusted their signals of sexual attrac-
tion based upon their perceived initial interest of
the other, and upon their own sexual attraction.
Women, who generally reported low levels of
sexual attraction actually inflated their signals.
4
Women also made more interrelated evaluations of their
partner’s short-term vs. long-term mate value than men,
particularly in Study1 (women: r.57 and r.31; men:
r.21 and r.17 for Study 1 and Study 2, respectively).
5
Correlations for the squared short-term and long-term
products were .64 and .50 for women and .38 and .24 for
men.
15ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Hence, our finding may explain Murray and
colleagues’ (2017) finding that women reported
having acted either less or more interested in
sex than they really were when dating someone
in the past.
The preceding findings of adjustment of sig-
nals of sexual attraction relative to self-reports
of felt attraction may partly reflect a coordina-
tion problem where payoffs are highest if you
seek to mirror the other party’s signals. Adjust-
ment of signals may prompt further communi-
cation between the parties, leaving more time
for interpretation and judgment. Strong attrac-
tion (more prevalent in men) combined with no
adjustment toward a party showing no interest
would most likely be quite intimidating and
off-putting (Bendixen & Kennair, 2015; Jona-
son & Li, 2013). Severe lack of attraction (more
typical in women), combined with no upward
adjustment of signals toward a more interes-
ted party, is likely to lead to being discarded and
abandoned for someone else before one has had
enough time to assess the other.
In the current study, participants were asked
about how strongly (along a 7-point response
scale) and what type (providing a qualitative
description) of sexual attraction signals they
sent. Signaling sexual attraction is merely one
of many possible ways of pursuing sexual ac-
cess. While the concept “pursuit” was not ex-
plicitly mentioned in the questionnaires, we be-
lieve sending signals of sexual attraction is a
fundamental aspect of sexual pursuit. Evidently,
the qualitative descriptions identified flirting,
seduction and solicitation behaviors for cues
that were categorized as moderate or strong by
independent raters. Moreover, this measure cor-
related substantially with strength of signals in
both studies. Consequently, it is likely that the
signals of sexual attraction measure has picked
up some form of pursuit.
Challenges Related to Self-Reported Sexual
Attraction
Is it possible that the reported low levels of
sexual attraction in women is due to a lower
ability recognizing sexual attraction, or that
they are less likely to report honestly about their
attraction following opposite sex encounters
compared to men? Possible evidence of the
former is found in a meta-analysis showing
weaker correlations between self-reported and
genital measures of sexual arousal for women
than for men (Chivers, Seto, Lalumière, Laan,
& Grimbos, 2010). Still, reporting one’s sexual
arousal may be both different from, and more
difficult than, reporting on one’s sexual attrac-
Figure 4. Level of own sexual attraction as a function of the short-term mate value of the
opposite sex partner. Scatter plots and predicted fitted values [95% CIs]. Panel A: Women
Study 1 (n137); Panel B: Women Study 2 (n134); Panel C: Men Study 1 (n82); Panel
D: Men Study 2 (n76).
16 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
tion. Hence, we find sex differences in arousal
recognition a less convincing explanation.
However, lack of recognition seems plausible
under two assumptions. First, that it is less risky
for a woman to appear to be oblivious to a
man’s sexual advances than to reject him ex-
plicitly. Second, following the logic of evolved
self-deception (Trivers, 2000), the most effec-
tive deception is the one that the deceiver be-
lieves, and that therefore a woman can, when
required, sidestep a man’s advances more effec-
tively if she is not fully aware of the degree of
her own sexual attraction. Also, one cannot
refute the possibility that women, relative to
men, are less likely to report sincerely on their
own sexual attraction and more likely to conceal
their true sexual motivation due to expected
societal negative reactions or some form of sex-
ual double standard (Kreager & Staff, 2009).
However, given the degree of anonymity during
responding (participants were instructed to keep
their responses private, and could return the
questionnaire in a prepaid envelope by mail)
and the gender similarity in level of reported
casual sex behavior, we do not anticipate dis-
honest or strongly biased responses due to so-
cial desirability issues in either sex. In addition,
in general, women participants admitted to sig-
naling greater sexual attraction than they felt.
The procedure and measurements applied to
our two studies provided a more direct test of
female sexual attraction and coyness than stud-
ies that have applied judgment of women’s sex-
ual intentions based on hypothetical behavior.
Coyness can express itself in one or both of two
ways: reporting less sexual attraction than one
feels and signaling less sexual attraction. We
relied on women’s reports to estimate both from
their most recent opposite sex encounter. The
possible functions and proximate causes of coy-
ness should result in predictable patterns in
these estimates. If coyness functions either to
avoid unwanted advances while reducing of-
fense, or to evade censure by daughter, sister, or
mate-guarding relatives and partners, then it is
more important to reduce signals of sexual at-
traction than to downplay felt attraction in later
(anonymous) reports to third parties. The over-
all findings from our two studies (including the
qualitative reports of signals sent) give us no
reason to assume that women’s reports of sig-
naled sexual attraction are biased or reflecting
coyness. At least not in the context of one of the
world’s most sexually liberal and gender egali-
tarian nations (Grøntvedt & Kennair, 2013).
Partnered participants’ sexual attraction and
signaling was markedly lower than single’s in
both men and women. This is consistent with
previous findings (Cole, Trope, & Balcetis,
2016). The validity of reports is further under-
scored by the analyses of other factors affecting
level of felt and signaled sexual attraction, such
as the level of perceived initial interest, and the
mate value of the opposite sex party.
Study Limitations and Future Research
This study used the naturalistic encounter
paradigm from earlier studies on misperception.
Further, it was expanded with more details of
what was perceived and signaled—and in Study
2, whether the outcome was sexual or not. As
such we believe the current findings improve
our understanding of the complex processes in-
volved in encounters between potential sexual
or romantic partners. However, we fully ac-
knowledge that there are several limitations in-
herent in this methodological approach. We ad-
dress a few of them here, but we strongly
recommend that future studies develop a di-
verse set of methods to discover the mecha-
nisms involved in the complex interplay in-
volved in encounters between potential
romantic partners.
The current results might not necessarily gen-
eralize to less gender egalitarian or less secular
societies with less sexual liberty. It might be
that the specific cultural context influences our
results, as women have less to lose in the cur-
rent culture. On the other hand, women from
this liberal culture are shown to worry more
than men about their sexual reputation follow-
ing casual sex (Kennair, Bendixen, & Buss,
2016). Still, level of sexual regret has been
found to reproduce cross-culturally, at least
across Western nations varying in relevant fea-
tures such as religiosity and sexual liberty (Ben-
dixen, Asao, Wyckoff, Buss, & Kennair, 2017).
Whether the current findings reproduce cross-
culturally needs to be examined in future studies
with samples from nations varying in gender
equality and sexual liberalism.
We also need to consider possible random
sample differences versus effects of context.
We chose freshmen/introduction weeks specif-
ically, because we wanted to be able to gather
17ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
enough reports about positive sexual outcomes;
our expectation was that there would be more
sex during this period. Further research needs to
consider this longitudinally: collecting a sample
to replicate the context features during freshmen
weeks, then follow up the same students several
months later in a more normalized context, and
finally test whether any changes mirror the find-
ings of the current study. Also, measuring ex-
perienced or subjective operational sex ratio
across the study period would provide more
insight into contextual factors affecting changes
in judgments of partner mate value and sexual
attraction reports in men and women.
An important caveat to consider when assess-
ing the findings using our methodology is the
high cognitive load put on the respondents an-
swering questions in retrospect on opposite sex
encounters that took part some (unknown to us)
time ago. However, memory may be facilitated
with shorter predefined reference periods (e.g.,
the last month), and researchers are advised to
time the last encounter by asking how many
days ago the encounter took place.
Further, any retrospective report could be dis-
torted by hindsight bias: For instance, single
people, or people who end up having sex may
retrospectively misinterpret their mate’s behav-
ior as indicating sexual interest. Still, at the
present we are not aware of any other method-
ology that could address coyness or playing
hard-to-get better without studying couples who
interact in experimentally controlled environ-
ments, which on the other hand introduces eco-
logical validity issues. Further, we are more
concerned about sex differences in mating strat-
egies than with the specific rates of the behav-
ior. If the rates of the actual behavior did not
differ, then we would still have to explain why
men and women differ in what they remember
more vividly. According to Conway and Pleydell-
Pearce (2000), one of the functions of autobio-
graphical memory is to make most memorable
those events that are relevant to the person’s mo-
tives. Sex and relationship status differences in
hindsight bias could then be plausibly linked to
differences in motives related to sexual strategies
for these groups. In signal detection theory, the
optimal location of the decision criterion is deter-
mined by base rates, sensitivity, and the subjective
evaluation of the possible outcomes. If our data
reflected retrieval biases driven by base rates, then
our original interpretation applies. If retrieval bi-
ases are driven by sex differences in the subjective
evaluation of the possible outcomes, those subjec-
tive evaluations would be motivations that influ-
ence the decision threshold, and thus behavioral
choices, which is the central claim of Error man-
agement theory (Galperin & Haselton, 2013). Sex
differences in sensitivity are not relevant to the
argument because low sensitivity does not gener-
ate bias, it only amplifies biases generated by base
rates and subjective evaluations (Lynn & Barrett,
2014). Therefore, sex differences in sensitivity
cannot explain sex differences that result in biases
on opposite sides of neutral.
Finally, although our measure of signaled
sexual attraction reported by the participants
may reflect some form of pursuit and level of
eagerness, we did not ask about pursuit or ea-
gerness directly in the questionnaire— only in-
directly through conveyed signals. These sig-
nals may not have been very costly, and future
studies ought to consider measuring these or
similar constructs more directly. This would
increase precision and provide a better test of
the male behavioral bias hypothesis.
Conclusion
Employing reported recall of naturally occur-
ring encounters with the opposite sex, we found,
in a normalized context, that men, but not
women, assumed higher levels of sexual interest
than the level of attraction that was reported by
the opposite sex. This provides some evidence
of male sexual overperception. During fresh-
man weeks, due to a substantial number of
freshmen women who reported being more sex-
ually attracted, this effect was not replicated in
Study 2. While men, in general, signal more
sexual attraction than women, men seem to cur-
tail signals of attraction toward women that they
perceived as having low sexual interest. There
was no indication that women downplay their
sexual signals relative to their own reported
sexual attraction. Rather, both men and women
were found to adjust their signals upward or
downward relative to their reported attraction,
possibly as a means to prompt further commu-
nication and to gain more information. There
were sex differences in what predicted who
ended up having sex. For men, a history of
sexual activity increases likelihood of having
sex; for women, how sexually attractive they
perceive their partner seems to be the primary
18 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
predictor. Despite limitations, the current study
advances our knowledge of the complex inter-
play between what one perceives, experiences
and signals in potential romantic meetings.
References
Bendixen, M. (2014). Evidence of systematic bias in
sexual over- and underperception of naturally oc-
curring events: A direct replication of Haselton
(2003) in a more gender-equal culture. Evolution-
ary Psychology, 12, 1004–1021. http://dx.doi.org/
10.1177/147470491401200510
Bendixen, M., Asao, K., Wyckoff, J. P., Buss, D. M.,
& Kennair, L. E. O. (2017). Sexual regret in U.S.
and Norway: Effects of culture and individual dif-
ferences in religiosity and mating strategy. Person-
ality and Individual Differences, 116, 246–251.
http://dx.doi.org/10.1016/j.paid.2017.04.054
Bendixen, M., & Kennair, L. E. O. (2015). Revisiting
judgment of strategic self-promotion and compet-
itor derogation tactics. Journal of Social and Per-
sonal Relationships, 32, 1056–1082. http://dx.doi
.org/10.1177/0265407514558959
Buss, D. M., & Schmitt, D. P. (1993). Sexual strat-
egies theory: An evolutionary perspective on hu-
man mating. Psychological Review, 100, 204–232.
http://dx.doi.org/10.1037/0033-295X.100.2.204
Chivers, M. L., Seto, M. C., Lalumière, M. L., Laan,
E., & Grimbos, T. (2010). Agreement of self-
reported and genital measures of sexual arousal in
men and women: A meta-analysis. Archives of
Sexual Behavior, 39, 5–56. http://dx.doi.org/10
.1007/s10508-009-9556-9
Cohen, J. (1988). Statistical power analysis for the
behavioral sciences (2nd ed.). Hilldale, NJ: Law-
rence Erlbaum.
Cole, S., Trope, Y., & Balcetis, E. (2016). In the eye
of the betrothed: Perceptual downgrading of attrac-
tive alternative romantic partners. Personality and
Social Psychology Bulletin, 42, 879– 892. http://dx
.doi.org/10.1177/0146167216646546
Conway, M. A., & Pleydell-Pearce, C. W. (2000).
The construction of autobiographical memories in
the self-memory system. Psychological Review,
107, 261–288. http://dx.doi.org/10.1037/0033-
295X.107.2.261
Darwin, C. (1871). The decent of man and selection
in relation to sex. London, UK: Murray.
Engeler, I., & Raghubir, P. (2018). Decomposing the
cross-sex misprediction bias of dating behaviors:
Do men overestimate or women underreport their
sexual intentions? Journal of Personality and So-
cial Psychology, 114, 95–109. http://dx.doi.org/10
.1037/pspi0000105
Galperin, A., & Haselton, M. G. (2013). Error man-
agement and the evolution of cognitive bias. In
J. P. Forgas, K. Fiedler, & C. Sedikides (Eds.),
Social thinking and interpersonal behavior (pp.
45– 64). New York, NY: Psychology Press.
Green, D. M., & Swets, J. A. (1966). Signal detection
and psychophysics. New York, NY: Wiley.
Grøntvedt, T. V., & Kennair, L. E. O. (2013). Age
preferences in a gender egalitarian society. Journal
of Social, Evolutionary, & Cultural Psychology, 7,
239–249. http://dx.doi.org/10.1037/h0099199
Hahn, A. C., Fisher, C. I., DeBruine, L. M., & Jones,
B. C. (2014). Sex ratio influences the motivational
salience of facial attractiveness. Biology Letters,
10, 20140148. Advance online publication. http://
dx.doi.org/10.1098/rsbl.2014.0148
Haselton, M. G., & Buss, D. M. (2000). Error man-
agement theory: A new perspective on biases in
cross-sex mind reading. Journal of Personality
and Social Psychology, 78, 81–91. http://dx.doi
.org/10.1037/0022-3514.78.1.81
Haselton, M. G., & Galperin, A. (2013). Error man-
agement in relationships. In J. A. Simpson & L.
Campbell (Eds.), Handbook of close relationships
(pp. 234–254): New York, NY: Oxford University
Press. http://dx.doi.org/10.1093/oxfordhb/978
0195398694.013.0011
Haselton, M. G., & Nettle, D. (2006). The paranoid
optimist: An integrative evolutionary model of
cognitive biases. Personality and Social Psychol-
ogy Review, 10, 47– 66. http://dx.doi.org/10.1207/
s15327957pspr1001_3
Jonason, P. K., & Li, N. P. (2013). Playing hard-to-
get: Manipulating one’s perceived availability as a
mate. European Journal of Personality, 27, 458–
469.
Kandrik, M., Jones, B. C., & DeBruine, L. M. (2015).
Scarcity of female mates predicts regional varia-
tion in men’s and women’s sociosexual orientation
across us states. Evolution and Human Behavior,
36, 206–210. http://dx.doi.org/10.1016/j.evolhum-
behav.2014.11.004
Kennair, L. E. O., Bendixen, M., & Buss, D. M.
(2016). Sexual regret: Tests of competing expla-
nations of sex differences. Evolutionary Psychol-
ogy, 14, 1–9. http://dx.doi.org/10.1177/1474
704916682903
Kennair, L. E. O., Wyckoff, J. P., Asao, K., Buss,
D. M., & Bendixen, M. (2018). Why do women
regret casual sex more than men do? Personality
and Individual Differences, 127, 61– 67. http://dx
.doi.org/10.1016/j.paid.2018.01.044
Kreager, D. A., & Staff, J. (2009). The sexual double
standard and adolescent peer acceptance. Social
Psychology Quarterly, 72, 143–164. http://dx.doi
.org/10.1177/019027250907200205
La France, B. H., Henningsen, D. D., Oates, A., &
Shaw, C. M. (2009). Social-sexual interactions?
Meta-analysis of sex differences in perceptions of
flirtatiousness, seductiveness, and promiscuous-
19ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
ness. Communication Monographs, 76, 263–285.
http://dx.doi.org/10.1080/03637750903074701
Lynn, S. K., & Barrett, L. F. (2014). “Utilizing”
signal detection theory. Psychological Science, 25,
1663–1673. http://dx.doi.org/10.1177/095679
7614541991
McHugh, M. L. (2012). Interrater reliability: The
kappa statistic. Biochemia Medica, 22, 276–282.
http://dx.doi.org/10.11613/BM.2012.031
McKay, R. T., & Efferson, C. (2010). The subtleties
of error management. Evolution and Human Be-
havior, 31, 309–319. http://dx.doi.org/10.1016/j
.evolhumbehav.2010.04.005
McNamara, J. M., Fromhage, L., Barta, Z., & Hous-
ton, A. I. (2009). The optimal coyness game. Pro-
ceedings Biological Sciences, 276, 953–960.
http://dx.doi.org/10.1098/rspb.2008.1273
Moss, J. H., & Maner, J. K. (2016). Biased sex ratios
influence fundamental aspects of human mating.
Personality and Social Psychology Bulletin, 42,
72– 80. http://dx.doi.org/10.1177/014616721
5612744
Murray, D. R., Murphy, S. C., von Hippel, W., Triv-
ers, R., & Haselton, M. G. (2017). A preregistered
study of competing predictions suggests that men
do overestimate women’s sexual intent. Psycho-
logical Science, 28, 253–255. http://dx.doi.org/10
.1177/0956797616675474
Okami, P., & Shackelford, T. K. (2001). Human sex
differences in sexual psychology and behavior.
Annual Review of Sex Research, 12, 186–241.
Penke, L., & Asendorpf, J. B. (2008). Beyond global
sociosexual orientations: A more differentiated
look at sociosexuality and its effects on courtship
and romantic relationships. Journal of Personality
and Social Psychology, 95, 1113–1135. http://dx
.doi.org/10.1037/0022-3514.95.5.1113
Perilloux, C., Fleischman, D. S., & Buss, D. M.
(2008). The daughter-guarding hypothesis: Paren-
tal influence on, and emotional reactions to, off-
spring’s mating behavior. Evolutionary Psychol-
ogy, 6. Advance online publication. http://dx.doi
.org/10.1177/147470490800600202
Perilloux, C., & Kurzban, R. (2015). Do men over-
perceive women’s sexual interest? Psychological
Science, 26, 70–77. http://dx.doi.org/10.1177/095
6797614555727
Regan, P. C. (1998). Minimum mate selection stan-
dards as a function of perceived mate value, rela-
tionship context, and gender. Journal of Psychol-
ogy & Human Sexuality, 10, 53–73. http://dx.doi
.org/10.1300/J056v10n01_04
Schmitt, D. P., & Buss, D. M. (1996). Strategic
self-promotion and competitor derogation: Sex and
context effects on the perceived effectiveness of
mate attraction tactics. Journal of Personality and
Social Psychology, 70, 1185–1204. http://dx.doi
.org/10.1037/0022-3514.70.6.1185
StataCorp. (2017). Stata statistical software: Release
15. College Station, TX: Author.
Swets, J. A., Dawes, R. M., & Monahan, J. (2000).
Psychological science can improve diagnostic de-
cisions. Psychological Science in the Public Inter-
est, 1, 1–26. http://dx.doi.org/10.1111/1529-1006
.001
Trivers, R. (1972). Parental investment and sexual
selection. In B. Campbell (Ed.), Sexual selection
and the descent of man, 1871–1971 (pp. 136–179).
Chicago, IL: Aldine.
Trivers, R. (2000). The elements of a scientific theory
of self-deception. Annals of the New York Acad-
emy of Sciences, 907, 114–131. http://dx.doi.org/
10.1111/j.1749-6632.2000.tb06619.x
Wachtmeister, C.-A., & Enquist, M. (1999). The
evolution of female coyness—Trading time for
information. Ethology, 105, 983–992. http://dx.doi
.org/10.1046/j.1439-0310.1999.00487.x
(Appendices follow)
20 BENDIXEN, KENNAIR, BIEGLER, AND HASELTON
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Appendix A
Means and SDs for Women and Men, Singles and Partnered, Study 1
Women (n141)
M(SD)
Men (n83)
M(SD)
Variable Single Partnered All Single Partnered All
Other’s Mate Value (ST) 3.56 (1.76) 2.97 (1.57) 3.25 (1.68) 5.04 (1.61) 4.68 (1.41) 4.89 (1.53
Other’s Mate Value (LT) 3.25 (1.61) 3.03 (1.41) 3.13 (1.51) 4.25 (1.39) 4.00 (1.46) 4.14 (1.42)
Own Mate Value (ST) 3.85 (1.47) 3.83 (1.55) 3.84 (1.51) 3.85 (1.68) 4.11 (1.21) 3.96 (1.49)
Own Mate Value (LT) 5.03 (1.15) 5.59 (0.92) 5.33 (1.06) 5.68 (1.11) 5.43 (1.01) 5.57 (1.07)
SOI-Behavior 3.19 (2.02) 2.32 (1.34) 2.72 (1.74) 3.23 (2.28) 2.50 (1.43) 2.92 (1.99)
SOI-Attitudes 5.09 (2.27) 4.65 (2.31) 4.85 (2.29) 6.04 (2.36) 6.24 (2.03) 6.13 (2.21)
SOI-Desire 4.34 (1.77) 2.22 (1.19) 3.21 (1.83) 5.59 (1.71) 4.14 (1.81) 4.97 (1.88)
Note.STShort-Term, LT Long-Term, SOI Sociosexual Inventory.
Appendix B
Means and SDs for Women and Men, Singles and Partnered, Study 2
Women (n135)
M(SD)
Men (n76)
M(SD)
Variable Single Partnered All Single Partnered All
Other’s Mate Value (ST) 4.57 (1.77) 3.47 (1.65) 4.04 (1.79) 4.91 (1.53) 4.37 (1.50) 4.72 (1.55)
Other’s Mate Value (LT) 4.07 (1.55) 3.00 (1.32) 3.56 (1.54) 4.16 (1.51) 3.37 (1.54) 3.83 (1.58)
Own Mate Value (ST) 4.15 (1.68) 3.89 (1.53) 4.03 (1.61) 3.62 (1.64) 4.07 (1.80) 3.80 (1.70)
Own Mate Value (LT) 4.82 (1.45) 5.50 (1.17) 5.14 (1.36) 5.16 (1.24) 5.87 (1.25) 5.45 (1.28)
SOI-Behavior 3.33 (1.95) 2.73 (1.51) 3.05 (1.77) 3.01 (2.13) 3.12 (1.97) 3.06 (2.04)
SOI-Attitudes 5.99 (2.34) 5.64 (2.18) 5.82 (2.26) 6.75 (1.82) 6.66 (2.09) 6.72 (1.91)
SOI-Desire 3.93 (1.90) 2.58 (1.15) 3.29 (1.73) 4.79 (1.88) 5.10 (1.98) 4.92 (1.90)
Note.STShort-Term, LT Long-Term, SOI Sociosexual Inventory.
Received July 31, 2018
Revision received February 4, 2019
Accepted February 4, 2019
21ADJUSTING SIGNALS OF SEXUAL INTEREST
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
  • ... Since nervous reactions are only part of the attraction and mate selection process, it may be the case that people aim to gather more information about the potential partner before letting their true feelings become evident. In support of this idea, it has been shown that both men and women tend to adjust their sexual interest signals upward or downward relative to their reported attraction to another, possibly as a means to prompt further communication in order to gain more information about that potential mate (Bendixen, Kennair, Biegler, & Haselton, 2019). It is also possible that those who mask their nervousness have had less romantic success, and/or are attempting to conceal the fact that their nervousness is a result of feeling they have a lower mate value than the potential mate. ...
    ... Therefore, these reactions may serve as an honest signal of interest when in the presence of a potential mate. It is also adaptive to attend to cues of romantic interest to either avoid or engage in expending energy on individuals who may or may not be interested or available (Bendixen et al., 2019;Floyd, Judd, & Hesse, 2008). ...
    Article
    Objective We sought to examine perceived nervous reactions when first interacting with a particularly attractive person/potential romantic partner. From a theoretical standpoint, we cogitate on the possible adaptive nature of these reactions that appear seemingly counterproductive for future mating success. Methods We documented 280 participants’ retrospective self-reports of experiencing a variety of physiological, vocal, and behavioral reactions during an initial encounter with a person they found highly attractive. We also asked participants to rate the reactions of others that they used to determine if another person was attracted to them. Results Participants reported most frequently experiencing increased attentiveness, smiling, staring, heart rate, giggling/laughter, blushing, and difficulty concentrating during this first encounter. Both sexes reported speaking faster and being less able to express themselves clearly, and women reported using a higher pitch and having a more unsteady tone of voice during an initial encounter of attraction. Further, participants reported observing similar nervous reactions by others whom they perceived were attracted to them. These findings were examined while considering individual differences in sociosexual orientation (i.e., propensity toward uncommitted sex), self-perceived mate value, empathy, and gender. Conclusions Participants reported that they did, indeed, experience a cluster of anxiety-related physiological and behavioral reactions during an initial encounter with someone they found highly attractive. Although appearing nervous and awkward during an initial encounter with an attractive other may seem counterproductive for future mating success, we discuss potential adaptive functions for displaying these responses.
  • Article
    Full-text available
    Sex differences in sexual regret are found to be robust across nations. Participants in Norway (N = 547) and the United States (N = 216) reported their level of regret for their most recent casual sexual experience. Participants also reported on proximate factors hypothesized to predict casual sex regret: negative emotions (worry), feeling pressured to have sex, low sexual gratification, partner's sexual competence, and to what extent they initiated the sexual encounter. Results suggest that greater feelings of worry, experiencing disgust, and feeling pressured to have sex predicted more regret for casual sex. In contrast, experiencing higher levels of sexual gratification, finding the partner to be sexually competent, and being the one taking the initiative predicted less sexual regret. Predictors of casual sex regret were not moderated by nation. However, relative to men, most of these predictors had a stronger impact on women's likelihood of regretting their most recent casual sex encounter. Discussion focuses on disgust, the key predictor of sexual regret, addresses limitations of the current study, and suggests future lines of research. Highlights •Six proximate factors related to casual sex regret examined. •Worry, disgust, and pressure predicted more regret. •Sexual gratification, sexual competency of partner, and taking initiative predicted less sexual regret. •Most predictors had a stronger impact on women's casual sex regret. •Predictors of casual sex regret were not moderated by nation.
  • Article
    Full-text available
    Men typically predict women’s sexual intentions to be higher than women say they are (Haselton & Buss, 2000). It is debated whether this cross-sex bias is because of men overestimating women’s intentions (Murray et al., 2017), women underreporting their own intentions (Perilloux & Kurzban, 2015, 2017), or both. To unify the current debate, we decompose the part of the bias attributable to women underreporting versus men overestimating by using a survey method intervention to reduce underreporting of sensitive information: eliciting estimates about others before sensitive self-reports. First, we calibrate the current measurement instrument to assess the overall size of the misprediction bias (Study 1). Then, we manipulate the order-of-elicitation of self- and other-reports (Studies 2 and 3): Women report significantly higher own sexual intentions when they are asked about other targets’ intentions before their own, suggesting that 48 to 69% of the overestimation bias is attributable to women underreporting their own sexual intentions. Analogous analyses for the misprediction bias about men suggest that women’s overestimation bias of men’s sexual intentions is entirely because of men underreporting their own sexual intentions. The findings have important implications for the current debate in the literature on cross-sex misprediction biases and the literature on asking sensitive survey questions.
  • Article
    Full-text available
    Sexual regret was investigated across two disparate cultures: Norway (N = 853), a highly secular and sexually liberal culture, and the United States (N = 466), a more religious and more sexually conservative culture. Sex differences, individual differences in preferred mating strategy, religiosity, and cultural differences in sexual regret were analyzed. Men were significantly less likely to regret having had casual sex than women and were significantly more likely to regret passing up casual sexual opportunities than women. Participants who were more religious regretted having had casual sex more and regretted passing up casual sex less. Sexually unrestricted participants were less likely to regret having had casual sex and were more likely to regret passing up casual sex. Finally, North Americans and Norwegians did not differ significantly in overall amount of sexual regret nor in patterns of sex differences in sexual regret. Discussion focuses the robustness of sex differences across cultures, the importance of explaining individual differences within cultures, and on future directions for cross-cultural research.
  • Article
    Full-text available
    The current study sought to answer three key questions about explaining the emotion of regret in the domain of casual sex: Are sex differences in sexual regret robust or attenuated in a highly egalitarian culture? What proximate psychological variables might explain sex differences in sexual regret? And what accounts for within-sex variation in experiences of sexual regret about casual sex. We conducted a study of 263 Norwegian students (ages 19–37) who reported how much they regretted having either engaged in, or passed up, their most recent casual sexual experience. Sex differences in sexual regret are not attenuated in this sexually egalitarian culture. The study revealed sex differences in worries about pregnancy, STIs, and reputation; however, these predictors did not succeed in accounting for the sex differences in regret engaging in casual sex. Sexual gratification and socio-sexual orientation both predicted the sex differences in casual sex regret. In contrast, only socio-sexual orientation attenuated the sex difference in regret passing up casual sex. Predictors of within-sex variation in casual sexual regret included worry about sexual reputation, experienced gratification during the encounter, and socio-sexual orientation. Discussion focuses on implications for the psychological design features of this relatively neglected emotion.
  • Article
    People in monogamous relationships can experience a conflict when they interact with an attractive individual. They may have a desire to romantically pursue the new person, while wanting to be faithful to their partner. How do people manage the threat that attractive alternatives present to their relationship goals? We suggest that one way people defend their relationships against attractive individuals is by perceiving the individual as less attractive. In two studies, using a novel visual matching paradigm, we found support for a perceptual downgrading effect. People in relationships perceived threatening attractive individuals as less attractive than did single participants. The effect was exacerbated among participants who were highly satisfied with their current relationships. The studies provide evidence for a perceptual bias that emerges to protect long-term goals. We discuss the findings within the context of a broader theory of motivated perception in the service of self-control.
  • Article
    Full-text available
    The operational sex ratio-the ratio of men to women in a given population-affects a range of social processes. The current research demonstrates that biased sex ratios (greater numbers of one sex than the other) influence fundamental aspects of people's mating strategy. When the sex ratio was favorable (one's own sex was in the minority), both sexes adopted strong sex-typical sociosexual orientations (relatively restricted for women; relatively unrestricted for men). When the sex ratio was unfavorable (one's own sex was in the majority), both sexes shifted toward the orientation typically favored by the other sex: Women became more unrestricted and men became more restricted (Experiment 1). When the sex ratio was unfavorable (relative to favorable), participants also displayed greater aggression toward a romantically desirable (but not undesirable) same-sex partner (Experiment 2). Exploratory analyses suggested that the sex ratio effect was present for unprovoked aggression but not provoked aggression (given the exploratory nature of that analysis, the aggression effect should be considered with some caution). Findings suggest that people's mating strategies are adaptively calibrated to contingencies within the local mating ecology.