Other people can tell whether your partner is cheating on you

Do humans have an infidelity radar?

We can identify a surprising amount of information about each other from the briefest of glimpses – a process that psychologists call thin-slicing. In the latest study in this area, a group led by Nathaniel Lambert have explored whether we can watch a romantic couple interact and tell within minutes whether one of them is a cheat.

Fifty-one student participants (35 women) in a relationship answered survey questions about their own infidelities toward their current partner. They and their partner were then filmed for three to five minutes performing a drawing task, in which one person is blindfolded and the other guides them as to what to draw.

Six trained coders (one man) later watched these clips and answered questions about whether the study participant in each couple had shown romantic interest in another person; flirted or made advances toward another person; or had sex with someone else. Answers to these questions were averaged to create an overall cheating verdict.

The coders’ cheating scores were correlated with the students’ self confessed levels of infidelity (the beta coefficient was .32; the researchers described the effect size as “moderate”). Further analysis showed this association was not simply due to the coders judging the participants’ social dominance, nor to them simply rating the male participants as more unfaithful on average. The researchers checked these possibilities because past research has linked social dominance with infidelity and because men are more often unfaithful than women.

Lambert’s team think these results show we’ve evolved a radar for spotting cheaters, an ability they think will have helped our ancestors to thrive, given the “adverse consequences of infidelity”. But what were the coders looking out for when they watched the videos?

A second study with 43 more undergrads was similar but this time the researchers also asked the coders to rate the participants’ commitment and trustworthiness. Again, the coders’ judgements of infidelity correlated with the students’ own admissions of having been unfaithful. Moreover, the coders’ judgments of infidelity were mediated by their verdicts about trustworthiness and commitment, so they seemed to be using inferences about these traits to inform their detection of cheating.

“Many people are interested in forming meaningful long-term romantic relationships and our research indicates that people may be internally programmed to identify inclinations that could be devastating to their relationship,” the researchers said. “Specifically, objective coders identified cheaters, and thus individuals seeking a committed relationship may be well advised to listen to their intuition or at least think twice before committing to someone they suspect may be inclined to cheat.”

Unfortunately, as well as being restricted to students and dating relationships, this research leaves many questions unanswered. We’re given little information about the coders, nor the training they received. Also, although we’re told the coders’ cheating judgments correlated with the students’ self-reported infidelity scores more than you’d expect if the coders were just guessing, it’s not possible from the available data to establish the rate of false alarms – those times that the coders felt a participant was a cheater when in fact they were not. You can imagine real life accusations based on such false alarms could cause a lot of emotional damage. Finally, the study unfortunately tells us nothing about exactly what behavioural cues (such as body language and tone of voice) the coders were using to make their judgments about infidelity.

_________________________________ ResearchBlogging.org

LAMBERT, N., MULDER, S., & FINCHAM, F. (2014). Thin slices of infidelity: Determining whether observers can pick out cheaters from a video clip interaction and what tips them off Personal Relationships DOI: 10.1111/pere.12052

Post written by Christian Jarrett (@psych_writer) for the BPS Research Digest.