Ultimate Mechanisms

Beyond humans, other socially “monogamous” animals also engage in sexual activity with conspecifics they are not pair-bonded to. A vast majority of birds form pair-bonds for the production and rearing of offspring, yet in more than 70% of species reviewed, at least some of the genetic offspring result from extra-pair copulations [6]–[7]. One extensive field study of the purportedly socially monogamous swift fox (Vulpes velox) reports that over 50% of females' offspring were genetically sired by males other than the partner to which she was pair-bonded [8]. And among primates, both males and females of the gibbon family (Hylobatidae) engage in extra-pair copulations despite strong pair-bonds [9]. Evolutionarily, such behaviors are presumably for the purposes of increased offspring diversity, increased offspring genetic quality, and/or extra fertilizations.

In humans, measures for global and national rates of promiscuity and infidelity are more varied. Worldwide median rates of non-paternity, or the rate of men raising children under the pretense of biological parentage, have been suggested to near 9% [10], or 10% more popularly [11]. However, this rate is mitigated by a male's degree of paternity certainty, with men who display high paternity confidence having a lower actual median non-paternity rate of 1.7% [11]. Further, an implicit condition of mate poaching, the active seduction of an individual away from a partner with whom they are romantically involved [3], generally includes infidelity on the part of the one being poached from their partner. Although mate poaching occurs as a “cultural universal” across no less than fifty-three previously studied nations [3], the frequency of mate poaching and all forms of uncommitted sexual behaviors vary widely among populations, reflecting differences in both local social constraints and individual behaviors [1]–[3], [5].

Individuals engaging in promiscuous sexual behaviors also face substantial risks, including diseases, unwanted pregnancies, and adverse social consequences. In addition, individuals engaging in infidelity also risk potentially losing or angering their pair-bonded partner. Historically and cross-culturally legal codes have responded to this evolutionary threat with biases accommodating entitlement to crimes of passion following evidence of adultery [12]. Despite these risks, sexual behavior in general and infidelity in particular offers enormous proximate and ultimate benefits in terms of immediate psychophysical reward and contributions to increased total number of offspring and subsequent increased diversity amongst those offspring. Given the potential disadvantages and advantages for social status and biological fitness, sex-related characteristics and sexual behavior itself are subject to substantial conflicting selective pressures [1]–[2], [5].

The substantial heritabilities between female monozygotic twins of 41% for number of lifetime sexual partners and 38% for infidelity demonstrates the contribution of genetics to these sexual behaviors [13]. Among monozygotic and dyzygotic twins of both gender, risky sexual behavior was correlated with personality traits and suggested heritability of 33% [14]. Further, high male reproductive variance enabled by promiscuity and infidelity of both sexes, irrespective of social monogamy, may be essential for the rapid fixation of beneficial genes within a population.

Proximate Mechanisms

Functionally, the brain's dopamine reward system supports motivation and libido to influence sexual and pair-bonding behaviors [15]–[17]. Dopaminergic function is enhanced by oxytocin and arginine vasopressin acting on the hypothalamus to regulate attachment [15], [17]–[20]. More specifically, it is oxytocin and vasopressin which most influence monogamy and pair-bonding [16], [18]–[20]. This is supported by recent genetic evidence of the vasopressin receptor 1a gene (AVPR1A) being associated with pair-bonding and marital satisfaction [21]. However, in a sample of female twins this same gene (AVPR1A) was not associated with infidelity or number of sexual partners [13]. Pair-bonded relationship satisfaction is not necessarily linked to promiscuity and infidelity, as the underlying motivations may be quiet different both psychosocially and neurobiologically. In fact, over two-thirds of adulterous men report that they would never have predicted their own infidelity [22]. Rather than envisioning relationship stability and infidelity as opposite ends of the same continuum, the underlying biological motivations for such behaviors may well be associated with distinct neurobiological pathways and thus more accurately thought of as orthogonal.

Additional studies from behavioral neurogenetics suggest that genes modulating dopamine neurotransmission mediate a variety of behavioral phenotypes associated with sensation-seeking [23]–[27]. One specific candidate is the dopamine D4 receptor (DRD4) gene, with a 48bp variable number tandem repeat (VNTR) polymorphism in exon III of chromosome 11. This polymorphic region typically includes 2 to 11 repeats. Individuals with at least one allele containing 7 or more repeats (7R+) show both reduced binding affinities and receptor densities for dopamine neurotransmission in the ascending corticomesolimbic reward pathway that extends from the ventral tegmental area to the nucleus accumbens, prefrontal cortex, and other cortical regions [28]. Individuals with these long alleles are predisposed to sensation-seeking behaviors including both migration and novelty-seeking, resulting in differential survival and reproduction [23], [25]–[26], [29]. The dopaminergic reward pathway influences physiological arousal, pleasure, and intrinsic reward [30]. Humans that possess at least one allele 7-repeats or longer (7R+) display behavioral phenotypes associated with attention deficit hyperactivity disorder (ADHD) [31], alcoholism [32]–[33], financial risk-taking [34], disinhibition and impulsivity [24], and sexual behavior [35], [36]. The latter includes associations between DRD4 genotype and sexual desire, arousal, and function [36], as well as likelihood of initiating sexual activity among young adults [35]. DRD4 7R+ has also been reported as unrelated to number of previous sexual partners [37], however the high reward-seeking variant of the dopamine transporter gene DAT1 was associated with increased number of sexual partners [38], although only in males. There is also support for an association between DRD4 7R+ and multiple/interracial ancestries [39]. DRD4 VNTR genotype frequencies vary widely across the globe, reflecting the adaptive utility of both motivations for dispersal [23] and in some cases culturally constrained behavioral phenotypes [26]. Taken together this suggests that the relationships between sexual behavior, human evolution, and dopamine modulating genes are fairly nuanced. Given the important role of dopamine in sexual behavior [17], it is predicted that variation in DRD4 will be associated with uncommitted sexual behavior in men and women.

Participants and Procedures

All aspects of this research were conducted in accordance with guidelines for the use of humans as research participants, and was approved by the University's AAHRPP accredited Human Subjects Research Review Committee. All participants provided written and verbal informed consent before initiating the study. Participants were 181 young adults (118 females, 63 males) recruited from a midsized state university in the northeastern United States. Average age of participants was 20.11 years (SD  = 1.62). Participants were primarily of European ancestry (61%), with 19% of Asian ancestry, 9% Hispanic, 1% African-American, 4% multi-racial, and 6% “Other.” Participants were recruited through the Psychology Department Human Subjects Research Pool and attended group assessment sessions seated every other seat and row with alternating order/color packets to ensure both actual and self-perceived privacy while they completed a confidential survey relating to sexual behavior and provided a DNA sample.

Phenotypic Assessment

The self-report survey included a comprehensive measure of demographics and questions on past sexual behavior, sexual expectations, and preferences, within which the primary dependent variables of interest were assessed. Participants were also assessed for smoking habits using the Fagerström Test of Nicotine Dependence (FTND) [40] and delayed reward discounting (DRD) using the Monetary Choice Questionnaire (MCQ) [41]. The MCQ is a self-report measure of DRD, which is an index of impulsivity, reflecting preferences for smaller immediate rewards compared to larger delayed rewards. In this case, participants made hypothetical choices for smaller amounts of money immediately versus larger amounts after a period of days (e.g., would you prefer $25 today, or $60 in 14 days); the MCQ provides indices of overall discounting as well as discounting for small, medium, and large rewards. Both the FTND and MCQ have been psychometrically validated.

Genotyping

For DNA analysis, a non-invasive oral buccal wash sample was obtained using 10 ml of Scope™ mouthwash [42]. Sufficient DNA for DRD4 PCR amplification was extracted from 96% (173/181) of the buccal cell samples. DNA was extracted using an abbreviated silica extraction protocol.

The human DRD4 gene on chromosome 11 contains a 48bp variable number tandem repeat (VNTR) polymorphism in exon III. Previous studies have highlighted problems associated with consistent genotyping of the DRD4 VNTR region [43], suggesting multiple PCR and electrophoresis runs for each sample to control for allelic dropout. Thus, the PCR reaction was modified to reflect the high GC content (see below) and all samples that were initially scored as homozygotes were reanalyzed two additional times with different starting template concentrations to unambiguously confirm genotypes. The PCR reaction consisted of 1x Q-Solution (Qiagen), 1x Buffer (Qiagen), 1 µM Primer 1 (5′ GCGACTACGTGGTCTACTCG 3′), 1 µM Primer 2 (5′ AGGACCCTCATGGCCTTG 3′), 200 µM dATP, 200 µM dTTP, 200 µM dCTP, 100 µM dITP, 100 µM dGTP, 0.3 units HotStar Taq (Qiagen), and 1 µl of DNA template, in a total volume of 10 µl. The PCR profile began with 15 minutes at 95°C for enzyme activation and denaturing of template DNA followed by 40 cycles consisting of 1 minute denaturation at 94°C, 1 minute annealing at 55°C, 1.5 minute extension at 72°C, and finished with a 10 minute extension at 72°C. Amplicons were electrophoresed through 1.4–2.0% agarose gels containing ethidium bromide and genotypes were determined by comparison with a 100 bp ladder.

Data Analysis

The primary dependent variables were history of sexual intercourse, infidelity (“cheated on” a committed partner), and promiscuity (“one-night stands”). In each case, categorical (i.e., Yes/No) and continuous (i.e., number) assessments were made. Nonparametric statistics and generalized linear model (GLM) analyses yielded largely parallel results, but for brevity only GLM results are shown when appropriate. Exact p values are reported for consideration by future studies. For continuous variables, medians with interquartile ranges are provided as measures of central tendency.

To facilitate truthful responding and per the Human Subjects Research Review Committee, participants were permitted not to answer individual items if they were uncomfortable given the personal nature of the assessment. This resulted in small portions of absent data: history of sexual intercourse, 8%; one night stands, 10%; infidelity, 4%. Genotypes were grouped as 7R+ (at least one allele 7-repeats or longer) or 7R- (both alleles less than 7-repeats); the 7R+ genotype was present in 24% of the sample. The analyses comprised three approaches, the primary approach and two strategies for ruling out alternative explanations. The primary analyses examined the relationship between the phenotypes of interest and DRD4 VNTR status. Second, possible contributions of population stratification were examined by re-running the analyses within the largest homogeneous group in terms of racial ancestry. Finally, the relationships among DRD4 VNTR genotype, sex, age, sexual behavior, smoking behavior, and impulsivity (DRD) were examined to determine the specificity of the associations in the primary analyses.

Sexual Behavior and DRD4 VNTR Genotype

Seventy seven percent of the sample reported a history of sexual intercourse, and no differences were evident between the genotypic groups (χ² [df  = 1]  = 0.32, p = 0.57). Among those who reported previous sexual experience, there was no genotypic difference in terms of the total number of sexual partners (F [1, 125]  = 1.59, p = 0.21). In contrast, categorical rates of promiscuous sex differed significantly between 7R+ and 7R- individuals (χ² [df  = 1]  = 5.58, p = .018). This difference reflected an almost 2-fold greater promiscuity rate in 7R+ (45%) compared to 7R- (24%) individuals (see Figure 1). However, no significant differences were evident in the total number of reported instances of promiscuity between the two genotypic groups among those who reported promiscuous sex (F [1], [33]  = .30, p  = .59). A similar pattern was evident for sexual fidelity, where 50% of 7R+ individuals reported being unfaithful compared to only 22% of 7R- individuals (See Figure 2), although this difference fell short of statistical significance (χ² [df  = 1]  = 2.15, p = .14). However, among individuals who reported sexual infidelity, 7R+ individuals reported significantly more extra-pair copulation partners (mean  = 1.79, SEM  = .20) compared to 7R- (mean  = 1.14, SEM  = .14) individuals (F [1], [40]  = 7.08, p = .011). Taken together, 7R+ were almost twice as likely to have engaged in promiscuous sex, and, when they were unfaithful, 7R+ individuals reported more than 50% more extra-pair copulation partners than 7R- individuals (See Figure 3).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Percent who report promiscuous sexual experiences, by DRD4 genotype group.

https://doi.org/10.1371/journal.pone.0014162.g001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Percent who report extra-relationship sexual experiences, by DRD4 genotype group.

https://doi.org/10.1371/journal.pone.0014162.g002

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 3. Number of extra-relationship sexual partners, by DRD4 genotype group.

https://doi.org/10.1371/journal.pone.0014162.g003

Sample Racial Ancestry Characteristics and Population Stratification

To examine possible population stratification effects (a “chopsticks gene”) [44], the preceding analyses were re-run in the largest racially homogenous ancestry (i.e., European, 61%), and all significant and nonsignificant findings were replicated. Participants were categorized racially according to their self-report of the racial/ethnic ancestry of all four of their grandparents. Group membership was assigned based on all four grandparents being of the same ancestry or being of multiple ancestries.

Within individuals of European ancestry, 7R+ individuals were more likely to report having engaged in promiscuous sexual behavior (χ² [df  = 1]  = 4.24, p = .04), but, among those who reported at least one promiscuous episode, no difference in the number of instances (F [1], [18]  = .21, p = .65). Also consistent with the primary findings, 7R+ individuals were no more likely to have been unfaithful (χ² [df  = 1]  = 1.60, p = .21), but, among those who reported infidelity, 7R+ reported a larger number of extra-pair copulation partners (F [1], [26]  = 5.83, p = .023). This is the same pattern as the total sample, suggesting systematic differences by racial/ethnic ancestry did not contribute to these findings.

Roles of Sex (Gender), Smoking Status, and Impulsivity

No sex differences were evident in terms of genotype frequency (female 7R+ frequency  = 23%, male 7R+ frequency  = 26%), suggesting it is unlikely that the findings are a function of systematic sex differences (χ² (df  = 1)  = .20, p = .66). Similarly, genotype frequency did not systematically vary by age (F [1, 167]  = 1.64, p  = .20) and age was not associated with continuous measures of promiscuity (Pearson r = .02, p = .90) or infidelity (Pearson r = .13, p = .40). These analyses suggest that participant age did not play a role in the primary findings. Genotype status was not associated with dichotomous smoking status (χ² [df  = 1]  = 1.12, p = .29). Interestingly, however, among cigarette smokers, 7R+ individuals had higher levels of nicotine dependence (mean  = 3.22, SEM  = .50) compared to 7R- (mean  = 1.47, SEM  = .34; F [1], [26]  = 8.32, p = .008). However, FTND was associated with neither instances of promiscuity (Pearson r = .02, p = .95), nor instances of infidelity (Pearson r = −.02, p = .97), suggesting independent relationships. Taken together, this suggests that the findings were not attributable to smoking status or nicotine dependence. Finally, DRD4 VNTR genotype was not associated with delayed reward discounting in general (F [1, 158]  = .60, p = .44), or for large (F [1, 158]  = .83, p = .37), medium (F [1, 158]  = .28, p = .60), or small (F [1, 158]  = .51, p = .48) rewards, further suggesting that the primary findings were not attributable to a more general preference for smaller immediate rewards over larger delayed rewards.