Accepted Manuscript

Title: Higher testosterone levels are associated with unfaithful
behavior in men

Authors: C. Klimas, U. Ehlert, T.J. Lacker, P. Waldvogel, A.
Walther

PII:

S0301-0511(18)30700-2

DOI:

https://doi.org/10.1016/j.biopsycho.2019.107730

Article Number:

107730

Reference:

BIOPSY 107730

To appear in:

Received date:

15 September 2018

Revised date:

22 May 2019

Accepted date:

10 July 2019

Please cite this article as: Klimas C, Ehlert U, Lacker TJ, Waldvogel P, Walther A,
Higher testosterone levels are associated with unfaithful behavior in men, Biological
Psychology (2019), https://doi.org/10.1016/j.biopsycho.2019.107730

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

1

Article Type:
Original research

Higher testosterone levels are associated with unfaithful

behavior in men

Running Head:
Testosterone levels in unfaithful men

Klimas, C.

1

, Ehlert, U.

2,3

, Lacker, T.J.

2,3

Waldvogel, P.

2

& Walther, A.

1,2,3

1

Biological Psychology, TU Dresden, Dresden, Germany

2

Clinical Psychology and Psychotherapy, University of Zurich, Zurich, Switzerland

3

Dynamics of Healthy Aging, University Research Priority Program, University of Zurich, Zurich,

Switzerland

Submitted to Journal:
Biological Psychology

Correspondence:
Dr. Andreas Walther
[e-mail]

Highlights



A robust relation between higher testosterone and increased unfaithful

behavior was observed.



Infidelity was measured using direct and sensitive indirect (crosswise)

questioning.



In this sample of men aged between 40 and 75 years, 37.5% men

answered having been unfaithful in the current relationship



Subsample analysis indicates a positive association between testosterone

and infidelity to be present primarily in men without sexual dysfunction.

Abstract

Background: Infidelity in romantic relationships is a common, but severe issue often causing

breakup and severe psychological impairment. Higher levels of testosterone are related to

ACCEPTED MANUSCRIPT

2

mating-behavior, sexual desire, and infidelity in men with sexual dysfunctions. Previous

studies, have insufficiently addressed the potential role of testosterone in infidelity in healthy

men.

Methods: A sample of 224 middle-aged self-reporting healthy men being currently in a

relationship completed questionnaires on relationship characteristics, infidelity, and provided

overnight-fasting saliva samples for testosterone quantification.

Results: In the sample, 37.5% men answered having been unfaithful in the current relationship,

while 29% were identified as fulfilling criteria for a sexual dysfunction. Adjusting for

covariates, a significant positive association for the frequency of unfaithful behavior and

testosterone levels emerged. Subsample analysis indicates a positive association between

testosterone and infidelity only to be present in men without sexual dysfunction.

Conclusion: Unfaithful behavior in males is associated with higher testosterone levels.

Key words:

Infidelity; testosterone; unfaithful behavior; sexual activity; sexuality; men; relationship

satisfaction; sexual dysfunction

Introduction

Infidelity is a severe problem in society and may lead to individual complaints like depression,

psychosomatic problems (Plack et al., 2008) and even symptoms of post-traumatic stress

ACCEPTED MANUSCRIPT

3

disorder (Gordon et al., 2004). Couples who had experienced infidelity before show less

relationship satisfaction and experience greater relationship problems than couples without

history of extrapair sexual activities (Plack et al., 2008). Therefore, infidelity threatens the

continuation of a relationship (Plack et al., 2008), making it one of the primary causes of divorce

(Amato and Previti, 2003; Scott et al., 2013).

Several studies have concluded that unfaithful behavior is more common in men

compared to women (Allen & Baucom, 2004; Haversath & Kröger, 2014; Laumann, Michael,

Gagnon, & Michaels, 1995; Wiederman, 1997). While some authors have identified trends for

infidelity rates becoming increasingly similar over time (Petersen & Hyde, 2010), others

observed higher rates in women compared to men, when applying a broad construct of

unfaithful behavior, which additionally accounted for kissing or romantic involvement, without

the necessity of sexual intercourse (Brand, Markey, Mills, & Hodges, 2007). However, when

examining exclusively sexual infidelity in large cohort studies, up to 15% of women, and 25%

of men admitted having had at least one sexual extradyadic activity in the current relationship

(Laumann, 1994; Wiederman, 1997). In a recently examined German sample, lifetime

prevalence of sexual infidelity even yielded 49% for heterosexual men and 29% for

heterosexual women, while the substantial gender difference was also present for homosexual

men (34%) and homosexual women (4%) (Haversath & Kröger, 2014).

From an evolutionary point of view, gender dissimilarities within the tendency of

extrapair sexual intercourse can be explained through the reproductive benefits for men when

having various short-term matings, potentially leading to a higher number of offspring

(Symons, 1979). This is supported by the fact, that men’s sperm is plentiful due to generation

of millions per day, which leads to high libido and enables the insemination of many females

(Buss & Schmitt, 1993). It was also shown that the sex hormone testosterone facilitates mating-

related behavior in terms of competitive behavior towards other males (Anestis, 2006; Gould

& Ziegler, 2007). Considering that men have on average six-fold higher testosterone levels than

ACCEPTED MANUSCRIPT

4

women (Clifton et al., 2016), and show considerably higher rates of unfaithful behavior than

women, higher testosterone levels might present an underlying biological parameter related to

more extradyadic sexual activities.

Supporting the mating effort hypothesis, research suggests that men who are involved

in a romantic relationship show 21% lower testosterone levels than single men (Burnham et al.,

2003), while fatherhood additionally lowers endogenous testosterone levels (Gettler et al.,

2011; Perini et al., 2012). Intriguingly, comparable effects were found in women (Barrett et

al., 2013; Kuzawa, Gettler, Huang, & McDade, 2010; van Anders, Hamilton, & Watson, 2007).

Men experiencing a reduced testosterone decline during the transition to fatherhood show

higher sexual activity at 4.5-year follow-up (Gettler, McDade, Agustin, Feranil, & Kuzawa,

2013), while higher testosterone in the phase of early fatherhood is associated with increased

perceived constraint due to fatherhood (Waldvogel & Ehlert, 2017). Interestingly, paired men

who still express great interest in extrapair sexual activities have higher testosterone levels than

those who see their relationship as exclusive (Edelstein, Chopik, & Kean, 2011; McIntyre et

al., 2006).

In previous studies, elevated testosterone levels in men and also women were associated

with higher sex drive in terms of seeking more frequent sexual encounters (Edelstein et al.,

2011; van Anders et al., 2007). For instance, polyamorous individuals show higher levels of

testosterone than monogamously partnered individuals (Alvergne, Faurie, & Raymond, 2009;

van Anders et al., 2007). In line with this, testosterone was shown to be positively associated

with sexual desire (Anderson, Bancroft, & Wu, 1992; Edelstein et al., 2011; McIntyre et al.,

2006).

Moreover, men suffering from hypogonadism show reduced libido and after exogenous

testosterone administration libido levels are significantly increased (e.g. Brock et al., 2016;

Corona et al., 2016; Cunningham et al., 2016, Wu et al., 2016), while the libido increasing

effect of testosterone has also been observed for men with normal testosterone levels (Anderson

ACCEPTED MANUSCRIPT

5

et al., 1992). Further, testosterone is positively linked to ejaculatory reflex (Corona et al., 2008)

and negatively associated with erectile dysfunction (Bandini et al., 2009; Walther et al., 2017).

Likewise, testosterone patches were shown to increase libido in women with hypoactive sexual

desire disorder (Braunstein et al., 2006). Thus, it has been shown that parenthood, sexual desire,

or relationship types are consistently associated with testosterone levels in males, while similar

associations in females are discussed. Besides this, testosterone administration studies

consistently lead to increased libido and sexual activity (Barrett et al., 2013; Braunstein et al.,

2006; Kuzawa et al., 2010; van Anders, Hamilton, & Watson, 2007).

Furthermore, research on personality traits and androgenization indicated positive

associations of testosterone with histrionic behavior like seductivity, self-centeredness and

sensation seeking (Bandini et al., 2009; Perini et al., 2012). Therefore, it can be assumed that

individuals with higher testosterone levels are more prone to extrapair sexual activities, as they

might easier experience monotony in monogamous relationships, but enjoy seduction of and

admiration by others.

Prior studies showed a positive association between testosterone and promiscuity, but

fell short for measuring infidelity with sufficient content validity (Edelstein et al., 2011;

McIntyre et al., 2006). In the study by Edelstein et al. (2011) examining 132 male and 122

female undergraduate students aged between 18 to 37 years, the infidelity measure consisted of

a subscale of a questionnaire measuring general promiscuity and degree of unrestricted sexual

orientation, while not explicitly asking about infidelity in the current relationship. McIntyre et

al. (2006) directly asked male undergraduate students about their history of extrapair sex, but

subsequently integrated this single item in the broader composite extrapair sexual interest

capturing also participants’ mere willingness to have extrapair sex. Fisher and colleagues

(2009) examining 2,592 heterosexual men reported significantly higher testosterone levels in

unfaithful men compared to faithful men. However, in this study, the sample consisted

exclusively of outpatients with sexual dysfunction, which has previously been shown to be

ACCEPTED MANUSCRIPT

6

negatively related to testosterone levels suggesting overall lower testosterone levels in this

sample than in the general male population. Unfaithful behavior was further identified

employing a face-to-face interview potentially increasing the risk of social desirability bias and

reducing infidelity reports. This is further supported by the relatively low infidelity rate of 17%

identified for this sample.

We therefore hypothesize that higher testosterone levels are associated with more

frequent unfaithful behavior and that the group of men reporting having engaged in extradyadic

sexual activities, shows higher testosterone levels compared to men having never engaged in

extradyadic sexual activities. In order to replicate and extend previous results, the current study

is the first to examine the relation between testosterone and infidelity in healthy men and men

identified as having a sexual dysfunction in parallel as well as including both hetero- and

homosexual men. Furthermore, we tested for infidelity through direct and indirect questioning

using for indirect questioning the cross-wise model – a questionnaire technique, which is

advantageous over anonymous self-reports to reduce risk of socially desirable answers –

thereby increasing the probability of a true and representative report of unfaithful behavior.

Methods

Participants

The examined sample represents a subsample of a larger project on men's health described in

detail elsewhere (Walther, Phillip, Lozza, & Ehlert, 2016). Of 224 men fulfilling the criterion

of being in a romantic relationship recruited via online and newspaper announcements for a

study on healthy aging, complete data on measures of infidelity and testosterone levels was

available. Since recent studies show no differences between homo- and heterosexual men with

regard to sex hormones (Juster et al., 2016), men of different sexual orientation were included

in the sample. The final sample comprised of 212 self-reporting (94.6%) heterosexual, 2 (0.9%)

ACCEPTED MANUSCRIPT

7

homosexual, 8 (3.6%) bisexual, and 2 (0.9%) asexual men. Sample characteristics for the 224

men aged between 40 and 75 years are represented in Table 1.

Measures and Procedures

After registering for the online platform and providing informed consent, participants

independently completed online psychometric test batteries one week prior to biological

examination. The psychometric test battery included sociodemographic information, the

International Index for Erectile Function (IIEF; Wiltink, Hauck, Phädayanon, Weidner, &

Beutel, 2003), the Relationship Assessment Scale (RAS; Sander & Böcker, 1993), the Brief

Symptom Inventory subscale on depressive symptoms (Zabora et al., 2001), the German

version of the Short Form Health Survey (SF36; Jenkinson et al., 1993), questions on alcohol

drinking habits, and questions on relationship characteristics. Further, we directly asked the

highly delicate question, whether and how often participants had ever engaged in extradyadic

sexual activities without agreement of the current partner with the following response options

“never”, “seldom”, “occasionally” and “regularly”. In the following, we classified participants

having responded either “seldom”, “occasionally”, and “regularly” to this question as

“unfaithful” and those having responded “never” as “faithful”.

We also used the Crosswise Model (Yu, Tian, & Tang, 2008), a sensitive indirect

questioning technique in order to reduce social desirability bias by providing an anonymous

response format (Jann, Jerke, & Krumpal, 2012), when assessing unfaithful behavior in the

past. Within the cross-wise model, participants are asked two simple yes-or-no-questions, a

non-sensitive one and a sensitive one. Participants then indicate whether the answer is “yes” to

both questions (option A), “no” to both questions (option A) or whether only one of the two

answers is true (option B). When looking at the responses of a single subject, the investigator

cannot conclude which of the two questions was affirmed or denied by the participant. Thus,

the Crosswise Model allows for participants´ anonymity and increases the likelihood of

ACCEPTED MANUSCRIPT

8

admitting norm violations. By considering the observed marginal distributions within the

overall sample, the proportion of subjects, who has responded with “yes” to the sensitive

question, can be estimated. Therefore, it is important that the probability of a potential yes-

answer to the non-sensitive question is known. In the present study participants were asked the

non-sensitive question in the first place: “Is your mother´s birthday in January or February?”.

The probability, that participants responded positively to the question was

2

12

= 0.167. The

sensitive question was asked in order to reveal unfaithful behavior of the participants:

“Considering your current relationship, have you ever had sexual contacts to a person outside

of this partnership without prior agreement of your partner?”. Subjects then provided a joint

answer for both questions. They chose option A when responses were identical and option B if

the answers differed. The proportion of unfaithful behavior was computed using the proposed

formula by Jann and colleagues (2012) shown below, with the unknown prevalence for

infidelity



, the observed proportion of respondents choosing answer option A,

𝜙̂, and the

known prevalence of a yes-answer to the non-sensitive question, p, in our study p = 0.167.

𝜋

̂

=

(𝜙̂ + 𝑝 − 1)

(2𝑝 − 1)

Procedures of testosterone sampling and quantification are described in detail elsewhere

(Walther et al., 2016). In brief, participants arrived at the biological examination after over-

night fasting and underwent different biological examinations (e.g. height and weight

measurement for the calculation of the body mass index). Saliva was sampled in Salicaps at

8:00 am under standardized conditions and testosterone was quantified by enzyme linked

immunosorbent assay in the biochemical laboratories of the Department of Psychology of the

University of Zurich. For testosterone, intra- and inter-assay coefficients were below 5% and

10%, respectively, while sensitivity was 1.8 pg/ml.

Statistical analyses

ACCEPTED MANUSCRIPT

9

To identify confounding variables being significantly different between the two groups

(unfaithful vs. faithful) two-tailed T-tests or χ

2

-tests were computed. To examine the relation

between the frequency of unfaithful behavior and testosterone levels one-tailed zero-order

bivariate correlations were calculated. Subsequently, analyses were repeated using one-tailed

partial correlations entering variables differing between the two groups or variables known to

influence salivary testosterone levels as covariates. Multiple linear regression analysis was

performed, predicting frequency of infidelity by testosterone levels and respective covariates.

Bias corrected and accelerated confidence intervals were reported on a 95%-level (BCa CI

95%

).

Binary logistic regression modeling was applied to predict infidelity as dichotomous variable

by testosterone levels and respective covariates. Analyses have been conducted with IBM SPSS

Statistics, Version 24.0. (Armonk, NY: IBM Corp.) significance level was set at alpha = .05.

Confounders

By using T-tests or χ

2

-tests, potentially differing variables between the self-reporting faithful

and the unfaithful group were examined (see Table 1). Identified potential confounders were

age, sexual function (IIEF), relationship satisfaction (RAS), depressive symptoms (BSI-18-D),

and alcohol consumption. Variables such as relationship duration or BMI were not significantly

different between groups.

Subsequently, a theoretical approach was used to further include potential variables

associated with salivary testosterone secretion. To control for the well-described effects of

general health, body composition, tobacco, recent sexual activity, or different medication types

on salivary testosterone secretion (Dai, Gutai, Kuller, & Cauley, 1988; English et al., 2001;

Walther & Ehlert, 2015), SF36 item 1 (How would you describe your general health status?

With answer options from “poor” to “excellent”), BMI, tobacco consumption, sexual activity

within the last 12 hours, and medication intake were included as covariates. To control for a

potential confounding effect of education, the highest degree of education ranging between “did

ACCEPTED MANUSCRIPT

10

not finish regular school” to “tertiary education” was included as covariate. To account for the

sampling method using saliva to extract testosterone, the potential contamination with blood by

gingival bleeding in the last days was controlled for (Kivlighan, Granger, & Schwartz, 2005).

Finally, to control for sleep restriction and sampling time variations due to participant delay to

examination appointments (Leproult, 2011), waking time of the participant and starting time of

saliva sampling were included as covariates.

Results

224 healthy men, who stated to be married or in a relationship at time of data collection, were

included in the analysis. Participants’ age ranged from 40 to 75 years (M = 58.05, SD = 10.69),

and mean relationship duration was 21.49 years (SD = 16.02). 84 participants (37.5%) reported

having engaged in extradyadic sexual activities one or more times without consent of the

current partner during their present relationship. When applying the crosswise model, 128 out

of 220 respondents (four participants did not agree to answer this specific question) chose

answer option A (

𝜙̂ = 0.58). After having applied the formula for estimated prevalence of

infidelity, results yielded a percentage of 37.7%, thereby verifying participants’ statements

about their sexual extrapair activities (see Table 2). By using the IIEF cut-off score for the

identification of sexual dysfunction as recommended by Wiltink et al. (2003), a total of 29%

men were identified as fulfilling criteria for at least a mild form of sexual dysfunction. Salivary

testosterone concentration ranged from 7.7 to 163.1 pg/ml (M = 66.42, SD = 26.37) indicating

comparable levels with other large male cohorts in the same age range reporting salivary

testosterone levels (Clifton et al., 2016).

One-tailed zero-order bivariate correlation between the frequency of unfaithful behavior

and testosterone levels yielded a positive trend (r = 0.095, p = .079). Subsequently conducted

partial correlation analyses, including identified covariates, revealed a significant positive

association between the frequency of unfaithful behavior and testosterone levels (r

p

= .14, BCa

ACCEPTED MANUSCRIPT

11

CI

95%

[.012, .280], p = .022). The calculated effect size for Cohen’s d = .28 suggests a small to

moderate effect (Kelley & Preacher, 2012). Point-biserial correlations between testosterone and

the dichotomous variable for infidelity (faithful = 0, unfaithful = 1) revealed a trend for a

positive association (r = .11, p = .057), indicating higher testosterone levels in unfaithful men.

This association became significant after controlling for respective covariates (r

p

= .17, BCa

CI

95%

[.030, .336], p = .007).

Multiple regression analyses were calculated, predicting frequency of extrapair sexual

activities through testosterone and respective covariates. Using the enter method it was

observed that the model explained a significant amount of variance in the frequency of

unfaithful behavior, F

15,207

= 2.11, p = .011, R

2

= .13, R

2

adjusted

= .07. The effect size of f

2

= .27

indicated a medium to large effect (Cohen, 1992). Testosterone significantly predicted the

frequency of extrapair sexual activities (β = .15, p = .045). The covariates alcohol consumption

(β = .17, p = .014), depression (β = .16, p = .037) and waking time (β = .14, p = .037) emerged

as additional significant covariates within the model. Further results as well as bias corrected

and accelerated confidence intervals are presented in table 3.

Results of the binary logistic regression analyses revealed that testosterone (b = .016, p

= .015) significantly predicted, whether participants were faithful or not, with higher

testosterone levels in unfaithful men. Also, alcohol consumption (b = .48; p = .001) and

depression (b = .48; p = .001) emerged as significant predictors.

Subsample analysis examining only men with identified sexual dysfunction and men

without sexual dysfunction is presented in the supplementary material. Although the reduction

of the sample sizes to 65 men (with sexual dysfunction) and 159 men (without sexual

dysfunction) leads to underpowered analyses, it emerges that healthy men are primarily

responsible for the significant positive association between testosterone and infidelity in the

entire sample (see supplementary material table 1a,b for correlational analysis as well as tables

2-5 for regression models).

ACCEPTED MANUSCRIPT

12

Discussion

Analyses revealed a positive association between testosterone levels and the frequency of

unfaithful behavior as well as infidelity operationalized as dichotomous variable (0 = not

unfaithful, 1 = unfaithful) in a self-reporting healthy sample of 224 hetero- und homosexual

men. This finding is line with previous research identifying for a large sample of 2,592 men

significantly higher testosterone levels in the 337 men reporting to be unfaithful (Fisher et al.,

2009). However, the entire sample by Fisher and colleagues (2009) consisted of Italian

heterosexual male outpatients with sexual dysfunctions, who had consulted a clinic for

andrology, which probably led to a selection bias towards a rather unhealthy sample since male

sexual dysfunction, particularly erectile dysfunction, has consistently been related to worse

general health (Guo et al., 2010). The here presented sample, by contrast, consisted only of self-

reporting healthy men, while it needs to be mentioned that also in the present sample 29% of

men were identified to have at least a mild form of a sexual dysfunction. Yet, this is a low

percentage for men in this age-group (Wagle, Carrejo, & Tan, 2012). Moreover, in the Fisher

et al. (2009) study unfaithful behavior was assessed through a face-to-face interview, which did

not allow for participants’ anonymity and probably led to an underestimation of infidelity in

the sample with an infidelity rate of 17%. By contrast, our analysis revealed sexual extrapair

activities in 37.7% for the current relationship (mean relationship duration 22 years), which is

more consistent with a recent anonymous internet survey in Germany reporting for heterosexual

and homosexual men a lifetime infidelity rate of 49% and 34%, respectively. Therefore, this

study is the first to report in a sample of self-reporting healthy men that increased testosterone

levels are associated with infidelity, while subsample analysis further indicates that this

association is driven by the portion of men of this sample, which were not suffering from any

sexual dysfunction (see supplementary material).

ACCEPTED MANUSCRIPT

13

Other studies also reported an association of testosterone and infidelity, while the

operationalized variable was often not infidelity but a related construct. For example, Edelstein

et al. (2011) assessed unfaithful behavior via the behavioral subscale of the Sociosexual

Orientation Inventory (SOI; Simpson and Gangestad, 1991), originally conceptualized for

measuring general promiscuity and degree of unrestricted sexual orientation (Penke &

Asendorpf, 2008). McIntyre and colleagues (2006), on the other hand, asked for participants’

history of extrapair sexual activities, but integrated this single item in the broader composite

extrapair sexual interest, capturing also participants’ mere willingness to have extrapair sex as

well as their individual SOI-score. Hence, in both studies, infidelity was confounded by general

sexual openness, thereby limiting content validity of the measurement. Associations between

higher testosterone and multiple partners are also reported by van Anders, Hamilton and

Watson (2007). Yet, the authors investigated monogamous vs. polyamorous individuals. The

latter responded to questions for people with two or more current partners and were mostly

recruited from polyamory groups. Subjects in long-term relationships were characterized as

monogamous, but were not surveyed about potential extradyadic sexual activities. We therefore

argue, that Anders and colleagues (van Anders et al., 2007) investigated the link between

testosterone and different types of romantic and sexual relations varying in their degree of

commitment and exclusiveness (i.e. monogamy vs. polygamy). But, there was no examination

of the relation between testosterone and infidelity per se. However, in line with our findings,

the studies mentioned above report higher testosterone levels to be associated with greater

desire towards extrapair sexual engagement and more frequent uncommitted sexual behavior

(Edelstein et al., 2011; McIntyre et al., 2006; van Anders et al., 2007). The identified association

between higher testosterone levels and increased frequency of unfaithful behavior is small to

moderate in effect size and thus in line with the broad literature. Infidelity is conceptualized as

a multifactorial condition. It was shown that relationship satisfaction (Mark, Janssen, &

Milhausen, 2011), or depressive symptomatology (Gorman & Blow, 2008) influence the risk

ACCEPTED MANUSCRIPT

14

for infidelity. This was further confirmed in our analysis showing lower levels of relationship

satisfaction and higher levels of depressive symptoms in unfaithful men (see table 1).

Interestingly, relationship satisfaction and depression are also related to testosterone levels

(Almeida et al., 2008; Feldman et al., 2002; Hooper et al., 2011). It is therefore assumed that

circulating testosterone levels in males stand in a mutually influencing relation with multiple

factors, while one such factor assembly consists of infidelity and associated constructs such as

promiscuity, willingness for extrapair activities, and engaging in polygamous relationships.

Men, naturally high in testosterone, tend to seek more frequent sexual encounters,

probably due to evolutionary reproductive benefits of having multiple sexual partners (Symons,

1979). Likewise, recent research has shown that men with high testosterone levels are perceived

as more attractive by women in terms of short-term mate judgments (O’Connor et al., 2014;

Roney et al., 2006). Research on personality traits and androgenization has shown that

testosterone levels are associated with histrionic behavior like seductivity (Bandini et al., 2009;

Perini et al., 2012) and narcissism (Pfattheicher, 2016). In line with this, high-testosterone

participants of our study may find it easier than low-testosterone subjects to approach potential

female sex partners because of the potentially more appealing impression they make on women.

In turn, this would offer greater opportunities for extrapair sexual activities.

Alternatively, the observed testosterone-infidelity-link might be mediated through

socioeconomical status (e.g. working position, income). Associations between higher

testosterone levels and higher status have been widely supported in previous literature

(Sherman, Lerner, Josephs, Renshon, & Gross, 2016), whereas elevated power in professionals

was shown to be associated with infidelity in both men and women (Lammers, Stoker, Jordan,

Pollmann, & Stapel, 2011). In turn, one could resume that the experienced power in men with

higher status increases their confidence in attracting potential sex partners. Likewise, men with

higher status and income may easily afford extrapair activities by hiring sex trade workers,

which men with lower income could not. However, statistics within a large sample of almost

ACCEPTED MANUSCRIPT

15

4,000 Mexican men about categories of extramarital sex partner demonstrated that only 5% had

extramarital sexual intercourse with a prostitute, whereas 54% reported extrapair sexual

activities with a friend, followed by the category mistress (17%), and coworker (14%)

(Pulerwitz, Izazola-Licea, & Gortmaker, 2001).

Positive associations between testosterone and sexual function were previously shown

in a large sample of middle-aged and older men (O’Connor et al., 2011). Interestingly, in the

present sample, unfaithful men, exhibiting higher testosterone levels, showed lower sexual

function scores compared to faithful men. Therefore, it is possible that unfaithful partners might

have experienced greater relationship distress before engaging in unfaithful behavior, which in

turn reduces relationship satisfaction and intimacy motivation (Allen et al., 2005; Treas &

Giesen, 2000). Both, relationship satisfaction and intimacy motivation, have been shown to be

important protective factors against sexual dysfunction in aging men (Walther et al., 2017).

Consistent with previous studies in this area (Fincham & May, 2017), in our study

relationship satisfaction was significantly different between groups, with lower satisfaction

among unfaithful men. Further, group comparisons revealed higher alcohol consumption in

unfaithful men, which is also in line with previous findings (Atkins, Yi, Baucom, &

Christensen, 2005; Graham, Negash, Lambert, & Fincham, 2016). Alcohol is considered a

testicular toxin (Maneesh, Dutta, Chakrabarti, & Vasudevan, 2006), leading to a decline of

plasma testosterone even in moderate drinkers (Sierksma et al., 2004). However, review articles

summarize that subjective sexual arousal increases after consumption of alcohol (Källmén &

Gustafson, 1998) and fosters sexual risk-taking (George & Stoner, 2000). Moreover, a

disinhibiting side effect of alcohol has been widely proposed (George & Stoner, 2000;

Gustafson & Källmén, 1989), while experimental studies show that the loosening of inhibition

through drinking relies on expectancies and self-fulfilling prophecies rather than on actual

pharmacological mechanisms (George, Stoner, Norris, Lopez, & Lehman, 2000). In sum, the

mentioned psychological consequences of alcohol probably makes frequent drinking men

ACCEPTED MANUSCRIPT

16

showing unfaithful behavior more often. An alternative explanation would be, that unfaithful

men tend to cope with alcohol consumption (Cook & Clark, 2005).

Findings in the present study do not allow for general conclusions about endogenous

testosterone levels predicting men’s disposition for infidelity. Recent research has suggested a

social modulation of androgenization rather than a simple hormonal causation (van Anders &

Watson, 2006). For example, testosterone levels increase as a direct consequence of sexual

arousal (Pirke, Kockott, & Dittmar, 1974; Stoléru, Ennaji, Cournot, & Spira, 1993),

masturbation (Purvis, Landgren, Cekan, & Diczfalusy, 1976) and sexual intercourse (Dabbs &

Mohammed, 1992). Moreover, findings from longitudinal studies show that masturbation

frequency predicts higher testosterone levels even five years later (Das & Sawin, 2016).

Furthermore, testosterone supplementation – being used by more and more men – increases

sexual desire and activity (Cunningham et al., 2016). This leads to the assumption that higher

testosterone levels in the unfaithful group might reflect a general effect of basal endogenous as

well as stimulated testosterone levels increasing the overall sexual desire and activity in males.

Moreover, unfaithful men in the present study showed lower relationship satisfaction,

which is often accompanied by diminished intimacy and lowered commitment to the long-term

partner. This may increase the urge to receive sexual satisfaction elsewhere. In line with this,

there is consistent evidence that those men still (or again) seeking potential partners, have

higher testosterone levels than men in a committed relationship (Alvergne, Faurie, & Raymond,

2009; Burnham et al., 2003; Gettler et al., 2013; van Anders et al., 2007). This underlines the

supporting effect of testosterone on male´s mating efforts and indicates that these efforts may

include short-term sexual activities namely extrapair sexual encounters.

Limitations

When interpreting these results, one needs to consider several limitations of this study. First,

although a relatively large sample of a homogenous group of men was examined post hoc power

ACCEPTED MANUSCRIPT

17

analysis based on identified effect size suggest that the present study was slightly

underpowered. Further, the cross-sectional nature of the study limits causal inferences. This

should be addressed in future studies applying experimental-longitudinal study designs

potentially examining the association between testosterone levels and infidelity prospectively

in men undergoing testosterone treatment. Further, although under standardized conditions, the

testosterone measurement was obtained at a single time-point being susceptible for short-term

fluctuations. Repeated sampling or long-term measurement of testosterone provided by

cumulative measures (e.g. hair testosterone) might be more adequate to reflect an individual’s

hormonal state. Though significant effects were identified, effect sizes are small to moderate

and a multi-factorial biopsychosocial model is suggested to underlie infidelity. Although there

are many findings that support an evolutionary explanation between testosterone and mating

effort (infidelity), we acknowledge that there are as many psychological and social theories that

explain a greater prevalence of infidelity in men. Due to our focus on the possible hormonal

component of infidelity in men, we have mainly discussed the evolutionary-biological

perspective, but do not deny the relevance of parallel psychological and social theories. Finally,

although infidelity was asked directly and calculated via the cross-wise measurement, the topic

is highly sensitive and skeptical participants still may have answered in a socially desirable

way.

Conclusion

Our findings indicate that males with higher testosterone levels show increased extrapair sexual

activity. Men identified as not having a sexual dysfunction are mainly responsible for this

significant association in the present sample. Data about unfaithful behavior of participants

have shown high reliability after verifying direct statements about frequency of extrapair sexual

activities via the Crosswise Model. Direct and indirect prevalence of infidelity among subjects

ACCEPTED MANUSCRIPT

18

were nearly equally high, indicating a low level of underreporting of infidelity in the present

study.

Although, dedicating to various sexual partners once had evolutionary benefits, in

today´s monogamous societies extrapair sexual activities often lead to severe relationship

conflicts thereby threatening not only the mental health and life quality of the betrayed partner,

but also that of the unfaithful partner. Higher testosterone levels may therefore be regarded as

potential risk factor for unfaithful behavior. This is especially important considering that

testosterone supplementation in men has received substantial attention for treating sexual

dysfunction or depression (Elliott et al., 2017; Walther et al., 2019) and an increasing number

of men worldwide is using testosterone products. Medical practitioners may thus consider to

inform men prior to using testosterone products about the possibility of an increased sex drive

due to an elevated basal testosterone level. This might be important to anticipate in exclusive

romantic relationships.

Author contributions

AW and UE designed the concept of the study. AW organized and conducted the study and

collected the data. CK and AW analyzed the data and wrote the first draft of the manuscript.

UE, TL, and PW contributed with important intellectual content and edited subsequent versions

of the manuscript.

Funding

This research was fully funded by the University Research Priority Program – Dynamics of

Healthy Aging of the University of Zurich.

Conflict of interest

ACCEPTED MANUSCRIPT

19

None declared.

Acknowledgements

We would like to thank all study participants for contributing to the study. Furthermore, we

would like to thank Thomas Bernauer for his support in implementing the crosswise

questioning procedure in the study design.

References

Allen, E. S., Atkins, D. C., Baucom, D. H., Snyder, D. K., Gordon, K. C., & Glass, S. P.

(2005). Intrapersonal, Interpersonal, and Contextual Factors in Engaging in and

Responding to Extramarital Involvement. Clinical Psychology: Science and Practice,

12(2), 101–130. doi: 10.1093/clipsy/bpi014

Allen, E. S., & Baucom, D. H. (2004). Adult Attachment and Patterns of Extradyadic

Involvement. Family Process, 43(4), 467–488. doi: 10.1111/j.1545-

5300.2004.00035.x

Almeida, O. P., Yeap, B. B., Hankey, G. J., Jamrozik, K., & Flicker, L. (2008). Low free

testosterone concentration as a potentially treatable cause of depressive symptoms in

older men. Archives of General Psychiatry, 65(3), 283–289. doi:

10.1001/archgenpsychiatry.2007.33

Alvergne, A., Faurie, C., & Raymond, M. (2009a). Variation in testosterone levels and male

reproductive effort: Insight from a polygynous human population. Hormones and

Behavior, 56(5), 491–497. doi: 10.1016/j.yhbeh.2009.07.013

Amato, P. R., & Previti, D. (2003). People’s reasons for divorcing gender, social class, the life

course, and adjustment. Journal of Family Issues, 24(5), 602–626. doi:

10.1177/0192513X03254507

Anderson, R. A., Bancroft, J., & Wu, F. C. (1992). The effects of exogenous testosterone on

ACCEPTED MANUSCRIPT

20

sexuality and mood of normal men. The Journal of Clinical Endocrinology &

Metabolism, 75(6), 1503–1507. doi: 10.1210/jcem.75.6.1464655

Anestis, S. F. (2006). Testosterone in juvenile and adolescent male chimpanzees (Pan

troglodytes): effects of dominance rank, aggression, and behavioral style. American

Journal of Physical Anthropology, 130(4), 536–545. doi: 10.1002/ajpa.20387

Atkins, D. C., Yi, J., Baucom, D. H., & Christensen, A. (2005). Infidelity in Couples Seeking

Marital Therapy. Journal of Family Psychology, 19(3), 470–473. doi: 10.1037/0893-

3200.19.3.470

Bandini, E., Corona, G., Ricca, V., Fisher, A. D., Lotti, F., Sforza, A., … Maggi, M. (2009).

Hysterical Traits Are Not from the Uterus but from the Testis: A Study in Men with

Sexual Dysfunction. The Journal of Sexual Medicine, 6(8), 2321–2331. doi:

10.1111/j.1743-6109.2009.01322.x

Barrett, E. S., Tran, V., Thurston, S., Jasienska, G., Furberg, A.-S., Ellison, P. T., & Thune, I.

(2013). Marriage and motherhood are associated with lower testosterone

concentrations in women. Hormones and Behavior, 63(1), 72–79. doi:

10.1016/j.yhbeh.2012.10.012

Brand, R. J., Markey, C. M., Mills, A., & Hodges, S. D. (2007). Sex Differences in Self-

reported Infidelity and its Correlates. Sex Roles, 57(1–2), 101–109. doi:

10.1007/s11199-007-9221-5

Braunstein, G. D., Sundwall, D. A., Katz, M., Shifren, J. L., Buster, J. E., Simon, J. A., …

Watts, N. B. (2006). Safety and Efficacy of a Testosterone Patch for the Treatment of

Hypoactive Sexual Desire Disorder in Surgically Menopausal Women: a Randomized,

Placebo-Controlled Trial. Journal of Urology, 175(2), 660–660. doi: 10.1016/S0022-

5347(05)00380-0

Burnham, T. C., Chapman, J. F., Gray, P. B., McIntyre, M. H., Lipson, S. F., & Ellison, P. T.

(2003). Men in committed, romantic relationships have lower testosterone. Hormones

ACCEPTED MANUSCRIPT

21

and Behavior, 44(2), 119–122. doi: 10.1016/S0018-506X(03)00125-9

Buss, D. M., & Schmitt, D. P. (1993). Sexual Strategies Theory: An evolutionary perspective

on human mating. Psychological Review, 100(2), 204–232. doi: 10.1037/0033-

295X.100.2.204

Clifton, S., Macdowall, W., Copas, A. J., Tanton, C., Keevil, B. G., Lee, D. M., … Wu, F. C.

W. (2016). Salivary testosterone levels and health status in men and women in the

British general population: Findings from the third national survey of sexual attitudes

and lifestyles (natsal-3). Journal of Clinical Endocrinology and Metabolism, 101(11),

3939–3951. doi: 10.1210/jc.2016-1669

Cohen, J. (1992). A power primer. Psychological Bulletin, 112(1), 155–159.

Cook, R. L., & Clark, D. B. (2005). Is there an association between alcohol consumption and

sexually transmitted diseases? A systematic review. Sexually Transmitted Diseases,

32(3), 156–164.

Corona, G., Jannini, E. A., Mannucci, E., Fisher, A. D., Lotti, F., Petrone, L., … Maggi, M.

(2008). Different Testosterone Levels Are Associated with Ejaculatory Dysfunction.

The Journal of Sexual Medicine, 5(8), 1991–1998. doi: 10.1111/j.1743-

6109.2008.00803.x

Cunningham, G. R., Stephens-Shields, A. J., Rosen, R. C., Wang, C., Bhasin, S., Matsumoto,

A. M., … Snyder, P. J. (2016). Testosterone Treatment and Sexual Function in Older

Men with Low Testosterone Levels. The Journal of Clinical Endocrinology &

Metabolism, (July), jc.2016-1645. doi: 10.1210/jc.2016-1645

Dabbs, J. M., & Mohammed, S. (1992). Male and female salivary testosterone concentrations

before and after sexual activity. Physiology & Behavior, 52(1), 195–197.

Dai, W. S., Gutai, J. P., Kuller, L. H., & Cauley, J. A. (1988). Cigarette smoking and serum

sex hormones in men. American Journal of Epidemiology, 128(4), 796–805.

Das, A., & Sawin, N. (2016). Social Modulation or Hormonal Causation? Linkages of

ACCEPTED MANUSCRIPT

22

Testosterone with Sexual Activity and Relationship Quality in a Nationally

Representative Longitudinal Sample of Older Adults. Archives of Sexual Behavior,

45(8), 2101–2115. doi: 10.1007/s10508-016-0815-2

Edelstein, R. S., Chopik, W. J., & Kean, E. L. (2011). Sociosexuality moderates the

association between testosterone and relationship status in men and women.

Hormones and Behavior, 60(3), 248–255. doi: 10.1016/j.yhbeh.2011.05.007

Elliott, J., Kelly, S. E., Millar, A. C., Peterson, J., Chen, L., Johnston, A., … Wells, G. A.

(2017). Testosterone therapy in hypogonadal men: a systematic review and network

meta-analysis. British Medical Journal, 7, 1–10. doi: 10.1136/bmjopen-2016-015284

English, K. M., Pugh, P. J., Parry, H., Scutt, N. E., Channer, K. S., & Jones, T. H. (2001).

Effect of cigarette smoking on levels of bioavailable testosterone in healthy men.

Clinical Science, 100(6), 661. doi: 10.1042/CS20010011

Feldman, H. a., Longcope, C., Derby, C. a., Johannes, C. B., Araujo, A. B., Coviello, A. D.,

… McKinlay, J. B. (2002). Age trends in the level of serum testosterone and other

hormones in middle-aged men: Longitudinal results from the Massachusetts Male

Aging Study. Journal of Clinical Endocrinology and Metabolism, 87(2), 589–598.

doi: 10.1210/jc.87.2.589

Fincham, F. D., & May, R. W. (2017). Infidelity in romantic relationships. Current Opinion

in Psychology, 13, 70–74. doi: 10.1016/j.copsyc.2016.03.008

Finkelstein, J. S., Lee, H., Burnett-Bowie, S.-A. M., Pallais, J. C., Yu, E. W., Borges, L. F.,

… Leder, B. Z. (2013). Gonadal steroids and body composition, strength, and sexual

function in men. The New England Journal of Medicine, 369, 1011–1022. doi:

10.1056/NEJMoa1206168

Fisher, A. D., Corona, G., Bandini, E., Mannucci, E., Lotti, F., Boddi, V., … Maggi, M.

(2009). Psychobiological Correlates of Extramarital Affairs and Differences between

Stable and Occasional Infidelity among Men with Sexual Dysfunctions. The Journal

ACCEPTED MANUSCRIPT

23

of Sexual Medicine, 6(3), 866–875. doi: 10.1111/j.1743-6109.2008.01140.x

George, W. H., & Stoner, S. A. (2000). Understanding Acute Alcohol Effects on Sexual

Behavior. Annual Review of Sex Research, 11(1), 92–124. doi:

10.1080/10532528.2000.10559785

George, W. H., Stoner, S. A., Norris, J., Lopez, P. A., & Lehman, G. L. (2000). Alcohol

expectancies and sexuality: a self-fulfilling prophecy analysis of dyadic perceptions

and behavior. Journal of Studies on Alcohol, 61(1), 168–176.

Gettler, L. T., McDade, T. W., Agustin, S. S., Feranil, A. B., & Kuzawa, C. W. (2013). Do

testosterone declines during the transition to marriage and fatherhood relate to men’s

sexual behavior? Evidence from the Philippines. Hormones and Behavior, 64(5), 755–

763. doi: 10.1016/j.yhbeh.2013.08.019

Gettler, L. T., McDade, T. W., Feranil, A. B., & Kuzawa, C. W. (2011). Longitudinal

evidence that fatherhood decreases testosterone in human males. Proceedings of the

National Academy of Sciences, 108(39), 16194–16199. doi: 10.1073/pnas.1105403108

Gordon, K. C., Baucom, D. H., & Snyder, D. K. (2004). An integrative intervention for

promoting recovery from extramarital affairs. Journal of Marital and Family Therapy,

30(2), 213–231. doi: 10.1111/j.1752-0606.2004.tb01235.x

Gorman, L., & Blow, A. (2008). Concurrent depression and infidelity: Background, strategies

for treatment, and future research. Journal of Couple & Relationship Therapy, 7(1),

39–58.

Gould, L., & Ziegler, T. E. (2007). Variation in fecal testosterone levels, inter-male

aggression, dominance rank and age during mating and post-mating periods in wild

adult male ring-tailed lemurs (Lemur catta). American Journal of Primatology, 69(12),

1325–1339. doi: 10.1002/ajp.20438

Graham, S. M., Negash, S., Lambert, N. M., & Fincham, F. D. (2016). Problem Drinking and

Extradyadic Sex in Young Adult Romantic Relationships. Journal of Social and

ACCEPTED MANUSCRIPT

24

Clinical Psychology, 35(2), 152–170. doi: 10.1521/jscp.2016.35.2.152

Guo, W., Liao, C., Zou, Y., Li, F., Li, T., Zhou, Q., … Mao, X. (2010). Erectile Dysfunction

and Risk of Clinical Cardiovascular Events: A Meta-Analysis of Seven Cohort

Studies. The Journal of Sexual Medicine, 7(8), 2805–2816. doi: 10.1111/j.1743-

6109.2010.01792.x

Gustafson, R., & Källmén, H. (1989). The effect of alcohol intoxication on primary and

secondary processes in male social drinkers. British Journal of Addiction, 84(12),

1507–1513.

Haversath, J., & Kröger, C. (2014). [Extradyadic Sex and its Predictors in Homo- and

Heterosexuals.]. Psychotherapie, Psychosomatik, Medizinische Psychologie, 64(12),

458–464. doi: 10.1055/s-0034-1390423

Hooper, A. E. C., Gangestad, S. W., Thompson, M. E., & Bryan, A. D. (2011). Testosterone

and romance: the association of testosterone with relationship commitment and

satisfaction in heterosexual men and women. American Journal of Human Biology,

23(4), 553–555.

Isidori, A. M., Giannetta, E., Gianfrilli, D., Greco, E. A., Bonifacio, V., Aversa, A., … Lenzi,

A. (2005). Effects of testosterone on sexual function in men: results of a meta-

analysis. Clinical Endocrinology, 63(4), 381–394. doi: 10.1111/j.1365-

2265.2005.02350.x

Jann, B., Jerke, J., & Krumpal, I. (2012). Asking Sensitive Questions Using the Crosswise

Model: An Experimental Survey Measuring Plagiarism. Public Opinion Quarterly,

76(1), 32–49. doi: 10.1093/poq/nfr036

Juster, R. P., Almeida, D., Cardoso, C., Raymond, C., Johnson, P. J., Pfaus, J. G., … Lupien,

S. J. (2016). Gonads and strife: Sex hormones vary according to sexual orientation for

women and stress indices for both sexes. Psychoneuroendocrinology, 72, 119–130.

doi: 10.1016/j.psyneuen.2016.06.011

ACCEPTED MANUSCRIPT

25

Källmén, H., & Gustafson, R. (1998). Alcohol and Disinhibition. European Addiction

Research, 4, 150–162.

Kelley, K., & Preacher, K. J. (2012). On effect size. Psychological Methods, 17(2), 137.

Kivlighan, K. T., Granger, D. A., & Schwartz, E. B. (2005). Blood contamination and the

measurement of salivary progesterone and estradiol. Hormones and Behavior, 47(3),

367–370. doi: 10.1016/j.yhbeh.2004.12.001

Kuzawa, C. W., Gettler, L. T., Huang, Y., & McDade, T. W. (2010). Mothers have lower

testosterone than non-mothers: Evidence from the Philippines. Hormones and

Behavior, 57(4–5), 441–447. doi: 10.1016/j.yhbeh.2010.01.014

Lammers, J., Stoker, J. I., Jordan, J., Pollmann, M., & Stapel, D. A. (2011). Power Increases

Infidelity Among Men and Women. Psychological Science, 22(9), 1191–1197. doi:

10.1177/0956797611416252

Laumann, E. O., Michael, R. T., Gagnon, J. H., & Michaels, S. (1995). The Social

Organization of Sexuality-Reply. JAMA: The Journal of the American Medical

Association, 274(7), 536–537. doi: 10.1001/jama.1995.03530070031020

Laumann, Edward O. (Ed.). (1994). The social organization of sexuality: sexual practices in

the United States. Chicago: University of Chicago Press.

Leproult, R., & E, V. C. (2011). Effect of 1 week of sleep restriction on testosterone levels in

young healthy men. JAMA, 305(21), 2173–2174.

Maneesh, M., Dutta, S., Chakrabarti, A., & Vasudevan, D. M. (2006). Alcohol abuse-duration

dependent decrease in plasma testosterone and antioxidants in males. Indian Journal

of Physiology and Pharmacology, 50(3), 291–296.

Mark, K. P., Janssen, E., & Milhausen, R. R. (2011). Infidelity in heterosexual couples:

Demographic, interpersonal, and personality-related predictors of extradyadic sex.

Archives of Sexual Behavior, 40(5), 971–982.

McIntyre, M., Gangestad, S. W., Gray, P. B., Chapman, J. F., Burnham, T. C., O’Rourke, M.

ACCEPTED MANUSCRIPT

26

T., & Thornhill, R. (2006). Romantic involvement often reduces men’s testosterone

levels--but not always: The moderating role of extrapair sexual interest. Journal of

Personality and Social Psychology, 91(4), 642–651. doi: 10.1037/0022-3514.91.4.642

O’Connor, D. B., Lee, D. M., Corona, G., Forti, G., Tajar, A., O’Neill, T. W., … the

European Male Ageing Study Group. (2011). The Relationships between Sex

Hormones and Sexual Function in Middle-Aged and Older European Men. The

Journal of Clinical Endocrinology & Metabolism, 96(10), E1577–E1587. doi:

10.1210/jc.2010-2216

O’Connor, J. J. M., Pisanski, K., Tigue, C. C., Fraccaro, P. J., & Feinberg, D. R. (2014).

Perceptions of infidelity risk predict women’s preferences for low male voice pitch in

short-term over long-term relationship contexts. Personality and Individual

Differences, 56, 73–77. doi: 10.1016/j.paid.2013.08.029

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(5), 1113–1135. doi:

10.1037/0022-3514.95.5.1113

Perini, T., Ditzen, B., Hengartner, M., & Ehlert, U. (2012). Sensation seeking in fathers: The

impact on testosterone and paternal investment. Hormones and Behavior, 61(2), 191–

195. doi: 10.1016/j.yhbeh.2011.12.004

Petersen, J. L., & Hyde, J. S. (2010). A meta-analytic review of research on gender

differences in sexuality, 1993–2007. Psychological Bulletin, 136(1), 21–38. doi:

10.1037/a0017504

Pfattheicher, S. (2016). Testosterone, cortisol and the Dark Triad: Narcissism (but not

Machiavellianism or psychopathy) is positively related to basal testosterone and

cortisol. Personality and Individual Differences, 97, 115–119. doi:

10.1016/j.paid.2016.03.015

ACCEPTED MANUSCRIPT

27

Pirke, K. M., Kockott, G., & Dittmar, F. (1974). Psychosexual stimulation and plasma

testosterone in man. Archives of Sexual Behavior, 3(6), 577–584. doi:

10.1007/BF01541140

Plack, K., Kröger, C., Hahlweg, K., & Klann, N. (2008). Außerpartnerschaftliche

beziehungen - Die individuelle belastung der partner und die partnerschaftliche

zufriedenheit nach dem erleben von untreue kurzbericht. Zeitschrift Fur Klinische

Psychologie Und Psychotherapie, 37(4), 231–235. doi: 10.1026/1616-3443.37.4.231

Pulerwitz, J., Izazola-Licea, J.-A., & Gortmaker, S. L. (2001). Extrarelational Sex Among

Mexican Men and Their Partners’ Risk of HIV and Other Sexually Transmitted

Diseases. American Journal of Public Health, 91(10), 1650–1652. doi:

10.2105/AJPH.91.10.1650

Purvis, K., Landgren, B. M., Cekan, Z., & Diczfalusy, E. (1976). Endocrine effects of

masturbation in men. The Journal of Endocrinology, 70(3), 439–444.

Roney, J. R., Hanson, K. N., Durante, K. M., & Maestripieri, D. (2006). Reading men’s faces:

women’s mate attractiveness judgments track men’s testosterone and interest in

infants. Proceedings of the Royal Society B: Biological Sciences, 273(1598), 2169–

2175. doi: 10.1098/rspb.2006.3569

Sander, J., & Böcker, S. (1993). Die Deutsche Form der Relationship Assessment Scale

(RAS): Eine kurze Skala zur Messung der Zufriedenheit in einer Partnerschaft.

Diagnostica.

Scott, S. B., Rhoades, G. K., Stanley, S. M., Allen, E. S., & Markman, H. J. (2013). Reasons

for Divorce and Recollections of Premarital Intervention: Implications for Improving

Relationship Education. Couple & Family Psychology, 2(2), 131–145. doi:

10.1037/a0032025

Sherman, G. D., Lerner, J. S., Josephs, R. A., Renshon, J., & Gross, J. J. (2016). The

interaction of testosterone and cortisol is associated with attained status in male

ACCEPTED MANUSCRIPT

28

executives. Journal of Personality and Social Psychology, 110(6), 921–929. doi:

10.1037/pspp0000063

Sierksma, A., Sarkola, T., Eriksson, C. J. P., Gaag, M. S., Grobbee, D. E., & Hendriks, H. F.

J. (2004). Effect of Moderate Alcohol Consumption on Plasma

Dehydroepiandrosterone Sulfate, Testosterone, and Estradiol Levels in Middle-Aged

Men and Postmenopausal Women: A Diet-Controlled Intervention Study. Alcoholism:

Clinical and Experimental Research, 28(5), 780–785. doi:

10.1097/01.ALC.0000125356.70824.81

Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality: evidence

for convergent and discriminant validity. Journal of Personality and Social

Psychology, 60(6), 870–883.

Stoléru, S. G., Ennaji, A., Cournot, A., & Spira, A. (1993). LH pulsatile secretion and

testosterone blood levels are influenced by sexual arousal in human males.

Psychoneuroendocrinology, 18(3), 205–218. doi: 10.1016/0306-4530(93)90005-6

Symons, D. (1979). The evolution of human sexuality. New York: Oxford University Press.

Treas, J., & Giesen, D. (2000). Sexual Infidelity Among Married and Cohabiting Americans.

Journal of Marriage and Family, 62(1), 48–60. doi: 10.1111/j.1741-

3737.2000.00048.x

van Anders, S. M., Hamilton, L. D., & Watson, N. V. (2007). Multiple partners are associated

with higher testosterone in North American men and women. Hormones and

Behavior, 51(3), 454–459. doi: 10.1016/j.yhbeh.2007.01.002

van Anders, S. M., & Watson, N. V. (2006). Social neuroendocrinology: Effects of social

contexts and behaviors on sex steroids in humans. Human Nature, 17(2), 212–237.

doi: 10.1007/s12110-006-1018-7

Wagle, K. C., Carrejo, M. H., & Tan, R. S. (2012). The Implications of Increasing Age on

Erectile Dysfunction. American Journal of Men’s Health, 6(4), 273–279. doi:

ACCEPTED MANUSCRIPT

29

10.1177/1557988311431629

Waldvogel, P., & Ehlert, U. (2017). Testosterone is Associated with Perceived Constraint in

Early Fatherhood. Adaptive Human Behavior and Physiology. doi: 10.1007/s40750-

017-0077-9

Walther, A., Breidenstein, J., & Miller, R. (2019). Association of Testosterone Treatment

with Alleviation of Depressive Symptoms in Men: A Systematic Review and Meta-

analysis. JAMA Psychiatry, 1;76(1), 31–40. doi: 10.1001/jamapsychiatry.2018.2734

Walther, A., & Ehlert, U. (2015). Steroid secretion and psychological well-being in men 40+.

In T. Rice & L. Sher (Eds.), Neurobiology of men’s mental health. New York, Nova.

(pp. 287–322).

Walther, A., Mahler, F., Debelak, R., & Ehlert, U. (2017). Psychobiological Protective

Factors Modifying the Association Between Age and Sexual Health in Men : Findings

From the Men ’ s Health 40 + Study. American Journal of Men’s Health, 11(3), 737–

747. doi: 10.1177/1557988316689238

Walther, A., Phillip, M., Lozza, N., & Ehlert, U. (2016). The rate of change in declining

steroid hormones : a new parameter of healthy aging in men ? Oncotarget, 7(38), 1–

28. doi: 10.18632/oncotarget.11752

Wiederman, M. W. (1997). Extramarital sex: Prevalence and correlates in a national survey.

Journal of Sex Research, 34(2), 167–174. doi: 10.1080/00224499709551881

Wiltink, J., Hauck, E. W., Phädayanon, M., Weidner, W., & Beutel, M. E. (2003). Validation

of the German version of the International Index of Erectile Function (IIEF) in

patients with erectile dysfunction, Peyronie’s disease and controls. International

Journal of Impotence Research, 15(3), 192–197. doi: 10.1038/sj.ijir.3900997

Yu, J.-W., Tian, G.-L., & Tang, M.-L. (2008). Two new models for survey sampling with

sensitive characteristic: design and analysis. Metrika, 67(3), 251.

Zabora, J., Brintzenhofeszoc, K., Jacobsen, P., Curbow, B., Piantadosi, S., Hooker, C., …

ACCEPTED MANUSCRIPT

30

Derogatis, L. (2001). A new psychosocial screening instrument for use with cancer

patients. Psychosomatics, 42(3), 241–246.

ACCEPTED MANUSCRIPT

31

Figure caption

Figure 1. Testosterone levels of faithful and unfaithful men. Values are presented as mean and

95% confidence intervals.

ACCEPTED MANUSCRIPT

32

Table 1. Descriptive statistics of the sample (N=224)

unfaithful

(N=84)

faithful

(N=140)

Characteristics

M

SD

M

SD

T

p

Age

59.80

10.62

57.01

10.62

1.903

.058

BMI

25.13

3.16

25.60

2.90

-1.126

.262

Sexual function (IIEF)

42.21

16.76

46.77

15.62

-2.057

.041

Health status (SF-36)

2.44

.61

2.39

.71

-.59

.555

Relationship duration

23.64

16.41

20.20

15.69

1.563

.120

Relationship

satisfaction (RAS)

4.05

.70

4.27

.67

2.241

.026

Depressive symptoms

(BSI-18-D)

1.87

2.87

1.13

2.34

-2.101

.037

Sexual activity (last

12 hours)

2.00

.00

1.99

.12

-1.42

.158

χ

2

p

Education level

5.25

1.17

5.21

1.36

6.883

.142

Tobacco consumption

1.40

.971

1.31

.897

3.164

.531

Alcohol consumption

3.94

1.09

3.34

1.09

18.813

.002

Medication intake

1.63

.485

1.68

.469

.531

.466

Note: BMI: body mass index; IIEF: International Index of Erectile Function; SF-36: Short Form Health Survey;
relationship duration in years; RAS: Relationship Assessment Scale; BSI-18-D: Brief Symptom Inventory 18 Items
Depression Subscale, tobacco consumption assessed by an item with a 5-point likert scale from “non-smoker” to
“smoking 10 or more cigarettes per day”, alcohol consumption on a 6-point likert scale from “I never drink
alcohol” to “2 or more alcoholic drinks per day; Medication intake assessed with “yes” or “no”.

ACCEPTED MANUSCRIPT

33

Table 2. Prevalence in % of infidelity via direct questioning and Crosswise-Model.

Direct Questioning

(DQ)

Crosswise-Model

(CM)

Difference

(CM-DQ)

infidelity

37.5

37.7

0.2

observations

224

220

Table 3. Linear model predicting frequency of extrapair sexual activity during the current
relationship y multiple predictors.

b

SE b

ß

p

Constant term

-.194

5.384

NA

.930

Testosterone levels (pg/mg)

.004 [.001, .007]

.002

.146

.045

Age

.007 [-.003, .017]

.006

.108

.213

Education level

-.020 [-.092, .053]

.036

-.038

.571

BMI

-.001 [-.034, .036]

.016

-.005

.943

Medication

.023 [-.166, .232]

.113

.016

.837

Cigarette consumption

.028 [-.054, .114]

.051

.038

.584

Alcohol consumption

.105 [.019, .203]

.042

.171

.014

General health status (SF36-1)

-.043 [-.194, .123]

.072

-.042

.552

Sexual function (IIEF-SF)

-.002 [-.009, .003]

.003

-.057

.458

Sexual activity

.561 [.211, .960]

.488

.077

.251

Depression (BSI-D)

.042 [-006, .093]

.020

.156

.037

Relationship satisfaction (RAS)

-.093 [-.276, .076]

.077

-.094

.225

Gingival bleeding

.105 [-.315, .499]

.187

.038

.574

Awakening time

.002 [.001, .003]

.001

.144

.037

Starting time of saliva sampling

-.117 [-1.284, .115]

.217

-.036

.589

Note. NA: not available. 95% bias corrected and accelerated confidence intervals reported in parentheses. Confidence
intervals and standard errors based on 1000 bootstrap samples. Explained variance of the frequency of extrapair sexual
activity by the model was 13% (R

2

= .13). BMI: body mass index, IIEF-SF: International Index of Erectile Function

– Sexual Function Scale; SF-36-1: Short Form Health Survey – Item 1; RAS: Relationship Assessment Scale; BSI-
18-D: Brief Symptom Inventory 18 Items Depression Subscale.

ACCEPTED MANUSCRIPT

34

Table 4. Binary logistic regression model predicting unfaithful vs. faithful men by multiple
predictors

b

SE b

95% CI for Odds Ratio

p

Lower

Odds

Upper

Constant term

-26.523

54413.032

NA

NA

NA

<.001

Testosterone levels

(pg/mg)

.016

.006

1.01

1.02

1.03

.015

Age

0.33

.019

.99

1.03

1.07

.081

Education level

.015

.123

.797

1.015

1.29

.904

BMI

-.014

.054

.89

.986

1.10

.801

Medication

.197

.372

.59

1.22

2.52

.597

Cigarette consumption

.097

.170

.79

1.10

1.54

.570

Alcohol consumption

.476

.146

1.21

1.61

2.14

.001

General health status

(SF36-1)

.000

.242

.62

1.00

1.61

.999

Sexual function (IIEF-SF)

-.014

.011

.97

.986

1.01

.212

Sexual activity

21.19

.511

NA

NA

NA

.999

Depression (BSI-D)

.112

.068

.98

1.12

1.28

.100

Relationship satisfaction

(RAS)

-.186

.250

.51

.83

1.35

.456

Gingival bleeding

.490

.636

.47

1.63

5.68

.441

Awakening time

.003

.003

.99

1.00

1.01

.188

Starting time of saliva

sampling

-2.86

.4.037

.001

.058

157.02

.479

Note. NA: not available. 95% confidence intervals for odds ratio. Explained variance of the frequency of extrapair sexual
activity by the model was 21% (R

2

= .21). BMI: body mass index, IIEF-SF: International Index of Erectile Function – Sexual

Function Scale; SF-36-1: Short Form Health Survey – Item 1; RAS: Relationship Assessment Scale; BSI-18-D: Brief
Symptom Inventory 18 Items Depression Subscale.

ACCEPTED MANUSCRIPT