“You Did it on Purpose, I Couldn’t Help It:” Differences in How We Explain Infidelity


Let’s pretend you run across your romantic partner holding someone else’s hand. What are your first thoughts? Do you think to yourself “there must be some explanation for this, maybe they are rehearsing for a play” or do you think “that sketchball is sneaking around behind my back”? Well, as it turns out, the way we interpret and perceive infidelity-related behaviors varies notably depending on the person in question.


According to a recent study of mine published in the journal of Basic and Applied Social Psychology (Thompson & O’Sullivan, 2017)people interpret the intentions of their romantic partner’s indiscretions very differently from the ways with which they interpret their own. This tendency to judge and interpret other’s behavior differently from or own is referred to as the actor-observer bias (Jones & Nisbett, 1972). In essence, social psychologists describe the actor-observer bias as the tendency to excuse our own negative behavior by attributing it to all sorts of circumstantial reasons, while explaining the negative behavior of others using stable dispositions (i.e., personality characteristics). For example, when explaining poor test performance, people are likely to attribute the cause of their own poor performance to situational variables (e.g., noisy test environment, a poor teacher/student relationship). On the other hand, people often attribute the negative test performance of others to dispositional variables (e.g., he/she lacks intelligence, he/she did not study enough). 

So, in my recent study, I was interested to see if these same biases applied to people’s attributions for infidelity. In particular, I wanted to know whether adults attributed the cause of their partner’s real-life indiscretions to dispositional variables to a greater extent than their own indiscretions and whether they attributed their own real-life indiscretions to situational variables to a greater extent that their partner’s indiscretions. Turns out that the results from 802 U.S. adults revealed just as I suspected: we are more likely to blame our partner’s infidelity on dispositional factors and our own infidelity on situational factors.

But that was not all, the data also indicated that the difference in how we explain our partner’s infidelity as compared to our own infidelity depended on the type of behavior. Using a 7-point scale (with higher scores indicating greater endorsement of dispositional attributions), adults attributed a partner’s real-life sexual infidelity (e.g., oral intercourse, vaginal intercourse, kissing) to dispositional features to a greater extent (M = 5.17, SD = 2.17) than one’s own real-life infidelity (M = 3.02, SD = 2.17). This same trend was true for technology-related infidelity (e.g., browsing singles website, sending a sexually-explicit text message, masturbating over webcam), with adults attributing a partner’s technology infidelity to disposition features to a greater extent (M = 4.85, SD = 1.87) than one’s own technology infidelity (M = 3.89, SD = 2.17). However, adults’ attributions for emotional (e.g., providing emotional support, sharing secrets, attending a formal event) and solitary infidelity (e.g., viewing pornography, engaging in masturbation alone, finding a celebrity attractive) did not vary between a partner’s or one’s own infidelity.

Inf Chart

So, why do we assume our partner’s infidelity is a result of dispositional characteristics and our own infidelity is a result of situational features? Some argue that this bias is self-serving and that we are motivated to protect our self-esteem by blaming our negative actions on situational variables that are beyond our control (e.g., our infidelity is less damaging if it was a result of intoxication rather than if it was a result of our own internal motives). However, this self-serving bias does not apply when judging others. We are less motivated to protect the self-esteem of others and, as a result, we don’t make the same attributions for the behavior of others as we do for ourselves.

Furthermore, why did we only see this actor-observer bias for certain forms of infidelity (primarily sexual and technology)? This may relate to the guilt we experience from engaging in certain forms of infidelity over others. In particular, the more discomfort/guilt produced by a behavior, the more likely one is to justify that behavior (i.e., make situational attributions) (for more information refer to Cognitive Dissonance Theory; Aronson, 1969). In fact, sexual infidelity and technology infidelity are the most extreme forms and likely produced the strongest feelings of discomfort/guilt, resulting in the use of more guilt-reducing strategies (Smith, 1961). Consequently, those who engaged in sexual or technology infidelity were likely to employ guilt-reducing strategies to a greater extent than were those who engaged in more benign behaviors, such as solitary infidelity. Furthermore, because people are unlikely to experience discomfort or guilt as a result of a partner’s infidelity, they are less likely to justify a partner’s infidelity behavior by making external attributions as they would their own. Thus, the actor-observer bias demonstrated in this research likely stems from the need to justify one’s own infidelity but not a partner’s.


Aronson, E. (1969). The theory of cognitive dissonance: A current perspective. Advances in Experimental Social Psychology, 4, 1-34.

Jones, E. E., & Nisbett, R. E. (1987). The actor and the observer: Divergent perceptions of the causes of behavior. In E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.), Attribution: Perceiving the causes of behavior (pp. 79-94). Hillsdale, NJ: Lawrence Erlbaum.

Smith, E. E. (1961). The power of dissonance techniques to change attitudes. Public Opinion Quarterly, 25, 626-639. doi: 10.1086/267058

Thompson A. E., & O’Sullivan L. F. (2017). Understanding differences in judgments of infidelity: An application of Attribution Theory. Basic and Applied Social Psychology. Advance online publication. doi:10.1080/01973533.2017.1350578