Mike Pence refuses to dine alone with women other than his wife. For the US vice president it is a mark of respect for his wife, Karen, and a rule guided by his strong religious convictions. Some commentators have hailed it the solution for men unable to control themselves, others call it patronising, sexist and insulting. (It is not an entirely rare attitude, however: in one study, around 5.7% of people surveyed considered buying someone of the opposite sex food an act of infidelity.)

Whatever you think about Pence’s justifications, at least he and Karen have clear boundaries about what is appropriate to do with people of the opposite sex – which is more than can be said for a lot of heterosexual couples.

Most people rarely have good definitions of exactly what it means to be unfaithful, and vastly underestimate how likely it is that some kind of betrayal will occur (despite being unfaithful themselves). They also have little understanding of how they will deal with infidelity if it does occur (with many people’s reactions surprising them).

Given its prevalence, that’s lack of communication and understanding is causing a lot of heartbreak – and many psychologists suggest that we should have much more open conversations about cheating.

Working out how many people have ever been unfaithful is challenging, not least because researchers are reliant on the honest confessions of cheaters. As a result, estimates of infidelity can vary wildly and are often affected by how data are collected. At the higher end of estimates, 75% of men and 68% of women admitted to cheating in some way, at some point, in a relationship (although, more up-to-date research from 2017 suggests that men and women are now engaging in infidelity at similar rates). One of the lowest published rates of infidelity is 14% – still a sizeable number.

Yet only 5% of people believe that their own partner had cheated or will cheat at some point in their relationship, meaning that even the most conservative estimates would suggest that this happens much more frequently than expected. Perhaps we’re too trusting of our partners.

“Those of us who are not depressed generally have a really inflated sense of how likely good things will happen and unduly low sense that bad things will happen,” says Susan Boon of the University of Calgary. “One possibility is that our low presumption that our partners will cheat on us is a manifestation of that. Alternatively, when you are in a relationship it might be helpful to have faith in your partner because it would be unhealthy to monitor their behaviour all the time.”

Here lies one of the issues; cheating means different things to different people. Researchers might pre-define what cheating constitutes to them, but everyone has a different interpretation, so interviewees might not agree with them.

“People overestimate the extent to which others approve of and engage in infidelity in relation to how much they do,” says Boon. “I’m not sure why people don’t talk about it considering how often you see it in movies or songs. Part of it is that we’re not aware of the variability of standards. We assume wrongly that what I consider unfaithful you would too. It also admits that maybe this could happen. People would prefer to believe that you wouldn’t do this.”

About 70% of people have not discussed with their partner what counts as cheating. Does downloading a dating app count, for example? Between 18% and 25% of Tinder users are in a committed relationship while using the dating app. Presumably, meeting up with people you met on Tinder does. Unsurprisingly, Tinder users who are already in relationships are more likely to have casual sex.

The people responding to the question about whether they thought their partner had ever been unfaithful were free to interpret infidelity in any way they chose. Perhaps that makes the 5% statistic even more surprising. For some people, cheating might only include sex, but for others, flirting with someone might count. With the freedom to interpret infidelity as we wish, we’re still very optmisitic that it will never happen to us.

Defining emotional infidelity is particularly difficult. One place where emotional transgressions might occur is in the workplace where overlapping professional and personal interests result in close relationships. Plausibly this would allow for opportunities to transgress from innocuous friendships to something more intimate.

In one study, researchers interviewed women about their attitudes towards workplace relationships. These women, all in their 30s and 40s and in committed relationships, were asked about times they felt the lines between appropriate and inappropriate workplace relationships became blurry.

“I can’t lie, I look forward to seeing him at work,” said one interviewee, “it feels like a stupid school girl, you know, like when you have a crush on somebody and you see them and you’re like ‘Oh!’ and you get excited.”

The interviewees concluded that physical intimacy is not necessary to elicit feelings of emotional infidelity. Withholding information, confiding in another, even thinking about the other person if it prevents you from thinking about your partner were enough. These are all things that might happen considering the amount of time we spend at work and the nature of forming close relationships with coworkers.

The interviewees talked about ‘relationship safeguarding’; predefining ground rules about what is and what is not appropriate. They also said that choosing to trust their partners was important for maintaining a healthy relationship. “And being in fitness, it can get physical just because [I’m] trying to show people how to do the correct workouts,” said another interviewee. “So, it was a conversation that we had to have... ahead of time just to say, ‘I’m going to trust you to do your job and it won’t go beyond that’.”

The behaviour of your partners friends can be enlightening as to their own attitudes about infidelity. The greater the proportion of your friends who you believe have cheated in their relationships, the more likely you are to have cheated in the past, and the more likely you are to say that you would be willing to cheat again in the future. We tend to surround ourselves with similarly adulterous, or non-adulterous, people.

It is clear that most people in monogamous relationships think that cheating is morally wrong. But, if someone has cheated, is the best course of action to admit guilt? When asked this question by researchers, people tend to say yes. In fact, more than 90% of people questioned say they would want to know if their partner has cheated on them.

One piece of research suggests that the importance of appearing loyal and pure is a key reason why people make those moral judgments. In fact, maintaining loyalty is more important than protecting someone’s feelings. If the most important thing was not to cause harm, then people would have said that keeping the affair secret is more ethical than confessing. Whether in reality this is the best course of action is another matter. Infidelity is the number one cause of divorce in the US.

Admitting to cheating is clearly going to hurt your partner's feelings – but there is a lot of variation in how people react. Greg Tortoriello, a psychologist at the University of Alabama has studied the effects of perceived failure on people; particularly, people whose personalities might mean they react poorly to failure. One example is narcissists, who seek the approval of others and are very conscious about how they present themselves.

“We assessed two types of narcissists: grandiose narcissists and vulnerable narcissists,” says Tortoriello. “A grandiose narcissist has an inflated sense of self-worth linked to higher self-esteem, whereas a vulnerable narcissist is sensitive to judgements from others and usually has lower self-esteem. In both cases, slight threats can activate aggressive behaviour.”

In one study by Tortoriello, participants imagined their partner was engaging in various types of infidelity. Some of the imaginary infidelities were based on emotional experiences; your partner talking late at night on the phone with another person and responds to their text rather than yours. Others were sexual.

“Grandiose narcissists wanted to assert power and control over their relationships when there was a threat of emotional infidelity,” says Tortoriello. “This took the form of verbal threats, physical threats, surveillance – remember these were hypothetical responses to imaginary situations. What we didn’t find is that those infidelity threats aroused more negative emotions.”

Vulnerable narcissists after emotional infidelity spent more time worrying and had more negative emotions. They took the infidelity personally.

In clinical terms, diagnoses of narcissism as a pathological disorder tend to be black and white – you are either a narcissist or you are not. Most behavioural psychologists like Tortoriello view narcissism as a sliding scale – everyone can be judged to have some of these qualities to a greater or lesser extent. In this study, he was specifically looking at people who were above average for these traits but who were not necessarily pathologically narcissistic.

“If you are in a relationship with one of these people and cheat sexually, it pretty much looks like they are trying to assert dominance and that will manifest in fairly destructive behaviours, but it gets more complicated with emotional infidelity,” says Tortoriello. “Vulnerable narcissists may not communicate to you that there are these concerns around their relationship and there is turmoil in the relationship. If I were to propose an intervention I would say finding ways to cultivate communication in specifically these relationships where there are a lot of internalised negative emotions is important.”

Forgiveness is most likely when cheating is an isolated incident and when an apology is offered. Though, Tortoriello and Boon reiterate that people react very differently in hypothetical situations and in reality. “Unanimously people say they would break up with someone for cheating but in reality it is not how people respond,” says Boon. “Sometimes it's the end of marriages but not always.”

Tortoriello has started to think about collecting real life data and is keen to explore the version of events from both sides of a couple. Do our partners think we’re being more unfaithful than we do? Do they see cheating where others see harmless flirting?

One thing to consider is that although the lifetime prevalence of infidelity is high – it will probably happen to many people at some point – the odds in any particular year are probably quite low. It doesn’t seem particularly pressing to talk about it right now.