This study examined whether individuals' definitions of infidelity differ based on perspective —evaluating their own behaviors versus their partner's identical behaviors — to test for the presence of self-serving bias in infidelity judgments. A nationally representative sample of 205 participants completed an online survey through Qualtrics. Participants were randomly assigned to evaluate potentially unfaithful behaviors from either a self-perspective ("You are exchanging inappropriate texts") or partner-perspective ("Your partner is exchanging inappropriate texts"). Behaviors were categorized into five factors through principal components analysis: active cheating, infatuation, social activities, fantasy, and discussion of sex life. Ratings were made on a 7-point scale from "definitely not cheating" to "definitely cheating." Independent-samples t-tests revealed significant perspective differences for three behavioral categories. As hypothesized, participants judged themselves more leniently than their partners for active cheating (p < .05) and social activities (p < .05). Unexpectedly, participants rated their own inappropriate discussion of sex life more severely than their partner's identical behavior (p < .001). No perspective differences emerged for infatuation or fantasy behaviors. Sex differences were found for active cheating and fantasy factors, with females rating behaviors as more unfaithful than males, but biological sex did not interact with perspective effects. Results provide mixed support for self-serving bias in infidelity judgments, suggesting that perspective-taking effects vary by behavior type. The reversal for discussing sex life may reflect unique concerns about violating partner privacy. These findings highlight the importance of explicit communication about relationship boundaries and have implications for couples therapy interventions addressing infidelity-related issues.

Keywords: infidelity, self-serving bias, relationship perspective-taking, emotional infidelity, physical infidelity, relationship boundaries, couples therapy, gender differences

Defining infidelity: variability and individual differences

Previous research has raised important questions about how infidelity is defined. More critically, researchers have examined how individual perceptions and attitudes shape whether specific behaviors are considered unfaithful. Several factors influence these perceptions. Individual differences, biological sex, past infidelity experiences, and religiosity all significantly impact how people define infidelity. These factors shape what behaviors people consider acceptable or unacceptable in romantic relationships. The current study extends this research by examining whether potentially unfaithful behaviors are viewed differently when performed by oneself versus one's partner.

Previous literature has demonstrated that no universal definitions of infidelity exist.¹ Many people argue that "cheating" consists primarily of physical contact. However, research shows that infidelity perceptions vary based on evolutionary theory. This theory suggests that men are more upset by their partner's physical infidelity, while women are more upset by their partner's emotional infidelity. Additionally, infidelity definitions may depend on personal feelings and behaviors that violate a partner's expectations. Past evidence has shown that personality traits play a role in defining infidelity. How individuals perceive potentially unfaithful acts predicts whether they consider these behaviors unfaithful.²

Cultural factors significantly influence how infidelity is defined and perceived across different societies. Cross-cultural research reveals substantial variation in what behaviors are considered unfaithful. For instance, cultures emphasizing collectivism versus individualism may differ in their tolerance for emotional versus physical infidelity.³ Religious traditions also shape infidelity definitions, with some cultures viewing pornography use or opposite-sex friendships as unfaithful, while others consider these behaviors acceptable.⁴

Gender role expectations vary across cultures, potentially influencing whether men and women are held to different standards regarding faithfulness. These cultural differences highlight the importance of considering broader social contexts when examining infidelity perceptions. However, most research on infidelity definitions has been conducted in Western, predominantly individualistic cultures, limiting our understanding of how these perceptions might differ in other cultural contexts.

Consequences of infidelity

Many couples in committed relationships have implicit or unspoken rules about infidelity. However, partners often disagree about what constitutes infidelity. This gap between what each partner considers acceptable can complicate relationships.⁵ Even when rules are explicitly stated, some behaviors might not be addressed or might remain ambiguous. Therefore, the complications mentioned earlier could still be problematic. Although no specific definition of infidelity exists, Fife et al.,⁶ suggest that infidelity is "a betrayal of this implied or stated commitment regarding intimate exclusivity. With infidelity, emotional and/or sexual intimacy is shared with someone outside of the primary relationship without the consent of the other partner (p.316)." However, this topic remains widely misunderstood. This misunderstanding can negatively impact relationships if left unaddressed in a specific relationship context. The understanding of infidelity is remarkably malleable. In a recent study, participants were asked to rate infidelity behaviors. They were also asked if their definition of infidelity had changed as a result of simply participating in the study. A significant number of participants (38%) reported that their definition of infidelity had been altered over the course of the survey.⁵ These findings suggest that infidelity has a wide variety of definitions. These definitions are not always shared by those involved in the same relationship. Furthermore, these definitions can change, even if they are initially shared.

Infidelity within a relationship has been associated with negative consequences for both psychological and physical health. It has also been a predictor of divorce and relationship dissolution.^7–9 Couples therapists have rated infidelity as the third most difficult problem for couples in therapy. They also consider it the second most damaging problem to couples' relationships, only behind physical abuse.¹⁰ Partners who engage in infidelity are also likely to experience psychological distress. In contrast, those who do not engage in infidelity have lower psychological distress (i.e., depression, shame, and guilt).

Religious beliefs have also been found to predict inconsistencies in judgments of infidelity. Religiosity has been associated with positive relationship outcomes. Individuals who are more religious have reported higher relationship quality and a lower likelihood of divorce.¹¹ However, research has also suggested that more religious people are more likely to report infidelity. For example, research has demonstrated that religious people are more likely to believe that pornography use is morally wrong. Some may perceive this as infidelity.^12,13 Other behaviors are ambiguous when aimed at an individual other than one's partner. These behaviors include hugging, talking on the phone, or receiving/giving gifts. These might be considered unfaithful by some individuals who are self-described as religious.^4,14

Potential role of self-serving bias in infidelity judgements

Given the previously described findings, opinions have remained divided on what constitutes emotional and physical infidelity. This is because specific acts may be ambiguous. Examples include going to a casual lunch with an ex-partner or watching pornographic material without a partner's knowledge. One notion that has been explored is that potential infidelity behaviors could be judged differently based upon whether it was done by oneself or one's partner. Individuals judge potentially emotionally unfaithful behaviors less harshly when they themselves take part in these behaviors. They judge more harshly when their partners take part in the same actions.¹⁵ Thompson & O'Sullivan¹⁶ also reported that participants judged their own behavior more permissively than their partner's. However, this was only true for emotional/affectionate and technology/online behaviors, not for sexual or solitary behaviors. These findings suggest that while a universal definition of infidelity doesn't exist, several factors influence what people consider infidelity. These factors include personality traits, biological sex, religiosity, and one's perspective (i.e. oneself versus one's partner).

Hypothesis

The current study examines whether one's definition of infidelity differs based on perspective. Specifically, it explores differences between considering one's own behaviors versus those of their partner. Given the previous findings, particularly those of Thompson & O'Sullivan and Thompson & O'Sullivan,^15,16 we hypothesized that individuals would be less likely to rate behaviors as constituting infidelity when done by themselves. In contrast, they would rate the same behaviors as more unfaithful when done by their partner. This would demonstrate a self-serving bias in defining infidelity. Addressing this hypothesis allows this study to further elaborate on the question of what constitutes infidelity and when certain behaviors are classified as unfaithful.

Participants and procedure

Participants were recruited through a program named Qualtrics which allows researchers to specify key demographic targets for the sample. Participants were recruited through Qualtrics' research panel service, which maintains a nationally representative database of potential participants. The platform uses stratified sampling techniques to ensure demographic representativeness across key variables including age, gender, race, education, and geographic region. In this case, the researchers specified that a national representative sample (in terms of age, gender, and race) of 200 participants should be targeted. Participants received dollar compensation and were not informed of the hypothesis of the study prior to their participation. Ages of the participants (n = 205) covered a range of 19 to 79, with the mean age of 42.28 years (SD = 16.33 years). 50% of participants identified as female (n = 102) and 50% as male (n = 103). For the category of race, 62% (n = 128) of respondents were White, 12% were Asian (n = 25), 11% were Black (n = 22), 9% were Latinx (n = 18), and 6% mixed were (n = 12).

Potential participants were sent a link to the survey through Qualtrics. After providing informed consent, participants completed the survey. They were free to skip any questions they were uncomfortable answering, and they could end their participation at any time without penalty of losing their compensation. Participants were also provided with the contact information of the lead researcher as well as the Institutional Review Board’s contact information for the institution where the study was housed. Data were analyzed using the Statistical Package for Social Sciences (SPSS, version 29).

The manipulation of the independent variable, one’s perspective on the behaviors in question (whether one’s self or their partner did the behavior) was included in instrument described in the following section. In the self-condition, prompts were worded, for example, as “You are exchanging inappropriate texts with another person.”

In the partner condition, the same prompt was worded as "Your partner is exchanging inappropriate texts with another person.” The dependent variables in the study were the five classifications of potential infidelity behaviors described below.

Instruments

Perceptions of infidelity: A full list of the questions used in the current study regarding beliefs about infidelity perceptions can be found in Appendix A (posed from the self-perspective). All items were rated on a scale from 1 (definitely not cheating) to 7 (definitely cheating) with the midpoint of 4 being neutral. A principal components factor analysis utilizing a promax rotation resulted in four factors with eigenvalues greater than 1. Items were considered to belong to a factor if they had a loading of at least .6 on the factor with loadings less than .4 on all other factors. An examination of the resulting pattern matrix demonstrated the presence of one factor corresponding to active cheating, a second factor corresponding to infatuation, a third factor corresponding to social activities, and a fourth factor corresponding to fantasy. A single item regarding the extent to which one inappropriately discusses their sex life was also examined because its pattern matrix loading (.885) suggested its presence as a stand-alone factor.

Active cheating: The active cheating factor was composed of 8 items that were averaged together to form a composite score. Example items include “You exchanged inappropriate texts with the person,” “You lie about your relationship status with the person,” and “You deleted secret messages exchanged with the person.” Cronbach’s alpha was .91, demonstrating excellent internal consistency.

Infatuation: The infatuation factor was composed of 3 items that were averaged together to form a composite score. The items were “You have a crush on the person,” “You fantasize about the person during the day,” and “You fantasize about the person during sex with your current partner.” Cronbach’s alpha was .88, demonstrating excellent internal consistency.

Social activities: The social activities factor was composed of 2 items that were averaged together to form a composite score. The items were “You go to a casual lunch with the person without your partner,” and “You like commenting on the person’s social media posts.” These items were strongly correlated (r = .54) suggesting that they form a single factor.

Fantasy: The fantasy factor was composed of 2 items that were averaged together to form a composite score. The items were “You follow sexually inappropriate accounts on social media,” and “You watch pornography without your partner knowing.” These items were strongly correlated (r = .65) suggesting that they form a single factor.

Discussion of sex life: As stated earlier, a single item (“You discuss your sex life inappropriately with the person”) was examined as a stand-alone factor due to its high loading of .885 while simultaneously demonstrating low loadings on the other factors previously described (all < .12).

Demographics: Participants were asked a series of questions regarding demographic information such as sex assigned at birth, age in years, ethnicity, sexual orientation, relationship status, and whether or not they had experienced infidelity (both as the cheater and the one cheated on) in a past relationship.

In order to test the main hypothesis of the study that participants would be more lenient regarding their own behaviors than those of their partner, a series of independent-samples t-tests were computed with perspective (self-versus partner) as the independent variable and the five categories of infidelity behaviors (active cheating, infatuation, social activities, fantasy, and discussion of sex life) as the dependent variables. One-sided tests were examined because the hypothesis specified a clear expectation for the direction of the effect. A significant difference of perspective on these dependent variables was demonstrated for three categories. For both active cheating, t(200) = -1.72, p < .05, one-sided, (self: M = 5.16, SD = 1.19, partner: M = 5.42, SD = .88) and social activity, t(204) = -1.86, p < .05, one-sided, (self: M = 3.12, SD = 1.51, partner: M = 3.52, SD = 1.60) the direction was in the hypothesized direction with more leniency being given to one’s own behaviors than those of their partner. For discussion of one’s sex life, t(203) = 3.68, p < .001, one-sided, (self: M = 4.79, SD = 1.80, partner: M = 3.82, SD = 2.00), however, the outcome was in the opposite direction than predicted with the partner’s behavior in this case being seen as more acceptable than one’s own actions. No differences were found for either the infatuation or fantasy factors when it comes to perspective taking (both ps > .05). These findings suggest that meaningful differences do exist when looking at what constitutes infidelity from one’s own perspective versus the perspective of one’s partner, though these differences might not always be in the expected direction.

As a follow-up, each of the tests reported here were re-run to examine potential interactions with biological sex given that this variable has been a reliable predictor of infidelity perceptions in the past. While sex differences did emerge for the active cheating and fantasy factors (both ps < .01) with females believing more of the behaviors to constitute cheating than males, in no case did sex interact with perspective to alter the pattern of the findings (all ps > .05). It is thus demonstrated that the effects of perspective taking reported here are independent of any influence of sex that might also be present.

The hypothesis of the current study was that individuals would view behaviors related to infidelity more severely when considering their partner doing them as opposed to carrying out the behaviors themselves. This hypothesis received mixed support. When considering active cheating and social activities, participants did indeed rate themselves more leniently than their partners. This is in line with the study’s hypothesis and past research demonstrating a double standard of sorts with reference to acceptable behaviors within relationships (Thompson & O’Sullivan, 2016; Wilson and O’Sullivan, 2017).^14,15

The finding that participants judged their own inappropriate discussion of sex life more severely than their partner's identical behavior represents a unique departure from typical self-serving bias patterns. This reversal has important theoretical implications. It suggests that self-serving bias may have boundaries, particularly when behaviors involve betraying another person's most private information. This finding aligns with research on privacy violations and trust, suggesting that people instinctively recognize the profound nature of exposing intimate details without consent (Petronio, 2002).¹⁷ Unlike other potentially unfaithful behaviors that primarily betray the relationship, discussing sex life details betrays the partner's personal privacy and dignity.

The key difference is that discussing the details of one’s sex life inherently involves exposing the other person in a way that casual lunches or even sexting don't. Sexting betrays the relationship, but discussing your sex life betrays your partner's most private information. People may instinctively recognize this as a deeper violation when they're the one doing the exposing. This could also explain why it's the one area where the usual self-serving bias flips. It is well known that self-serving bias plays a significant role in one’s thought processes, with people typically attributing their successes to internal factors like talent or effort, while blaming external factors like luck or other people for their failures to maintain self-esteem. However, close relationships could place limits on an individual's self enhancement tendencies, making individuals more modest when in a close relationship.

Applications

The current study helps to shed further light on the ways in which people define infidelity within the confines of a close relationship, an area that, as previously noted, can vary widely from one person to another. One application of the current findings is that the members of a relationship dyad should discuss these issues early in their relationship to be sure that they are on the same page when it comes to what behaviors are acceptable versus off limits. Coming to an early agreement on these issues could improve relationship quality down the line by helping to avoid conflict over behaviors such as going on a casual lunch date that are ambiguous and could be interpreted negatively by one’s partner. If questions arise during the course of the relationship concerning whether a behavior is acceptable or not, it should be brought promptly to one’s partner for discussion so that potential problems can be avoided or resolved.

A second application of these findings is that they can be used in clinical or therapeutic settings as a starting point for infidelity-related interventions. Discussing issues related to infidelity definitions could open new channels of communication between members of existing couples or could shed light on why previous relationships have not worked out adequately for those seeking assistance with understanding relationship difficulties who are currently single. Exploring these issues in a controlled setting such as therapy led by a trained clinician could prove beneficial for both couples and single individuals.

From a clinical perspective, the finding that participants judged their own inappropriate discussion of sex life more severely than their partner's identical behavior has important implications for couples therapy. Therapists should be aware that clients may feel particularly guilty about sharing intimate details with others, even more so than engaging in other potentially unfaithful behaviors. This guilt may be justified, as such disclosures can feel like a fundamental violation of trust that extends beyond the relationship itself. Therapeutic interventions might focus on helping couples establish clear boundaries about what aspects of their intimate life can be discussed with others. Additionally, when addressing infidelity involving intimate disclosures, therapists should recognize that the offending partner may already feel heightened shame, which could be leveraged therapeutically for relationship repair. This behavior category may also reveal gender differences worth exploring, as men and women may have different comfort levels with intimate disclosure and different reactions to their partner's disclosures.

Future research should address several important limitations of the current study while building on its contributions. First, methodological improvements are needed to increase ecological validity. While hypothetical scenarios provide controlled conditions for testing, they may not reflect how individuals would respond to actual infidelity situations. Future studies could employ experience sampling methods to capture real-time reactions to potentially unfaithful behaviors, or utilize longitudinal designs to track how infidelity perceptions change over time and relationship stages. Additionally, experimental paradigms that create more realistic decision-making scenarios could provide stronger tests of self-serving bias in infidelity judgments.

Second, research should expand to diverse populations and relationship structures. Cross-cultural replication studies are critically needed to determine whether self-serving bias in infidelity judgments is universal or culturally specific. Studies examining infidelity perceptions in LGBTQ+ relationships, polyamorous arrangements, and other non-traditional relationship structures would broaden our understanding of how relationship context influences these judgments. Age-specific research could also reveal whether infidelity perceptions change across developmental stages, particularly as digital communication becomes increasingly integrated into relationships.

Social desirability issues should also be considered when interpreting these findings. It’s likely that at least some of the participants were overly conservative when responding to the prompts, not wanting to appear as if “anything goes” with regard to potential infidelity behaviors. This effect was likely stronger in the self than the partner condition, in line with the self-serving bias which says that people want to appear as socially acceptable as possible. Future research should consider including a social desirability scale in order to partial out these potential confounding effects.

The generalizability of these findings may be limited by cultural factors. This study was conducted with a U.S. sample, and infidelity perceptions may vary significantly across cultures with different values regarding relationships, gender roles, and individual autonomy. Future research should examine these patterns in diverse cultural contexts to determine whether self-serving bias in infidelity judgments is universal or culturally specific.

Finally, clinical applications warrant further investigation. Intervention studies could test whether communication-based approaches that explicitly address perspective-taking differences improve relationship satisfaction and reduce infidelity-related conflicts. Development of standardized assessment tools for couples therapy could help therapists identify areas where partners have divergent infidelity definitions, enabling more targeted interventions. Such applied research would translate these findings into practical tools for relationship enhancement and infidelity prevention.

In closing, the current work has shed important light on the role that one’s perspective in a relationship has in helping to shape how infidelity is defined. While more exploration is needed to clarify some of the shortcomings of this study, it is believed that the findings presented here lay the groundwork for this future work to be successful.

None.

None.

The authors declare that they have no conflicts of interest relevant to this manuscript.

1. Blow AJ, Hartnett K. Infidelity in committed relationships: A methodological review. J Marital Fam Ther. 2005;31(2):183–216.
2. Thornton V, Nagurney A. What is infidelity? Perceptions based on biological sex and personality. Psychol Res Behav Manag. 2011;4:51–58.
3. Zandbergen DL, Brown SG. Culture and gender differences in romantic jealousy. Pers Individ Differ. 2015;72:122–127.
4. Mattingly BA, Wilson K, Weidler DJ. Foggy faithfulness: relationship quality, religiosity, and the perceptions of dating infidelity scale in an adult sample. J Fam Issues. 2010;31(11):1465–1480.
5. Schonian S. Perceptions and definition of infidelity: A multimethod study. [UNLV Theses, Dissertations, Professional Papers, and Capstones.] 2013.
6. Fife ST, Weeks GR, Gambescia N. Treating infidelity: An integrative approach. Fam J. 2008;16(4):316–323.
7. Amato PR, Rogers SJ. A longitudinal study of marital problems and subsequent divorce. J Marriage Fam. 1997;59(3):612–624.
8. Negash S, Cui M, Fincham FD, et al. Extradyadic involvement and relationship dissolution in heterosexual women university students. Arch Sex Behav. 2014;43(3):531–539.
9. Previti D, Amato PR. Is infidelity a cause or consequence of poor marital quality? J Soc Pers Relat. 2004;21(2):217–230.
10. Whisman MA, Dixon AE, Johnson B. Therapists’ perspectives of couple problems and treatment issues in couple therapy. J Fam Psychol. 1997;11(3):361–366.
11. Mahoney E, Krumrei EJ, Pargament KI. Broken vows: Divorce as a spiritual trauma and its implications for growth an decline. In S. Joseph & P. A. Linley (Eds.), Trauma, recovery, and growth: Positive psychological perspectives on pasttraumatic stress (pp. 105-123). John Wiley & Sons, Inc. 2008.
12. Perry SL. Pornography use and marital quality: Testing the moral incongruence hypothesis. Pers Relat. 2018;25(2):233–248.
13. Zitzman ST, Butler MH. Wives’ experience of husbands’ pornography use and concomitant deception as an attachment threat in the adult pair-bond relationship. Sex Addict Compulsivity. 2009;16(3):210–240.
14. Wilson K, Mattingly BA, Clark EM, et al. The gray area: exploring attitudes toward infidelity and he development of the Perceptions of Dating Infidelity Scale. J Soc Psychol. 2011;151(1):63–86.
15. Thompson AE, O’Sullivan LF. I can but you can’t: Inconsistencies in judgments of and experiences with infidelity. J Relatsh Res. 2016;7.
16. Thompson AE, O’Sullivan LF. Understanding variations in judgments of infidelity: an application of attribution theory. Basic Appl Soc Psychol. 2017;39(5):262–276.
17. Petronio S. Boundaries of privacy: Dialectics of disclosure. State University of New York Press. 2002.