by Jordann L. Brandner, Gary L. Brase, & Jadyn Pohlman
Picture this: you catch the eye of an attractive person across the room, and you go to speak with them. They smile, they laugh at your jokes, and they lean in to listen to your conversation. Would you be able to say if they are sexually interested in you?
Our recent research published in Evolution and Human Behavior asks just that – can men and women reliably tell if someone is sexually interested in them or not? Previous research suggests that men overperceive these situations, often mistaking a woman’s friendliness as cues of sexual interest. Error Management Theory explains this “male overperception” effect as an evolved bias – it is less costly to the man to accidentally overperceive a woman’s interest than it is to miss a mating opportunity.
However, previous research on this overperception have relied on difference scores measures, which oversimplify the situation. In fact, men could have a tendency to respond in a particular way, regardless of it being right or wrong (consistently overperceiving as suggested by Error Management Theory), or alternatively, they could be answering based on how distinct the signals of attraction are from the signals of rejection. In more theoretical terms, evolutionary optimality could be reached by always assuming attraction, but it could also be reached by consistently correct observations, as these correct observations could also avoid the costly error of missing out on someone who was attracted to them.
This possibility was tested using Signal Detection Theory. This theory provides both a measure of bias (how strongly a person prefers to answer “interest” or “disinterest”), and also a measure of sensitivity, which tells how different signals are from non-signals for each person. Moreover, these measures are standardized, allowing comparison across different biases, and are individualized, allowing us to test if mating-relevant personality traits affect sensitivity, bias, both, or neither.
In Study 1, undergraduates were shown written scenarios similar to the one above, then asked if this person was sexually interested in them, how much interest the person had in them, and also answered some personality questions. Study 2 asked online participants similar questions, but using scenarios with much more ambiguous – and even conflicting – behaviors. Importantly, each scenario was pre-judged to determine if it signaled sexual attraction or not by both men and women, and how much attraction was signaled by it. This provided an independent evaluation for comparison.
Contrary to previous research, we actually found underperception of sexual interest – people in our studies had a preference for answering that the person was not interested. Moreover, women were overperceiving interest while men were underperceiving it – completely opposite to what we had predicted. That said, this bias rarely came into play, as participants in both studies were highly accurate at telling if a person was interested in them or not, even with the intentionally confusing behaviors used in Study 2. Traits such as life history strategy, sociosexual orientation, and mate value did not affect participants’ sensitivities or biases.
Contrary to previous research, we actually found underperception of sexual interest – people in our studies had a preference for answering that the person was not interested. Moreover, women were overperceiving interest while men were underperceiving it – completely opposite to what we had predicted.
Interestingly, analyzing these same data with the more typical difference scores measures did not yield the same results. While this analysis did similarly find underperception overall, it also appeared to support an overperception effect for men, despite men’s perceptions being closer to pre-rated interest levels. To understand how such a result could appear imagine people tasked with driving a certain speed, say, 40 miles per hour. Everyone is going slower than that, but men are going 38 mph whereas women are going 35 mph. Men are going relatively fast (analogous to overperception), but they are actually closer to the target speed.
Men’s difference between the pre-rated levels and their perceptions was smaller, and a smaller difference when negative is a larger number, resulting in the conclusion that men perceived “more” interest than women, when they were simply closer to the truth than women.
Our research combined the powerful and well-established measures from Signal Detection Theory with the evolutionary application from Error Management Theory. The Signal Detection Theory approach incorporating both bias and sensitivity is a better option than the traditional difference score approach. This method reflects the raw data better and allows for more precise, comparable results. It also raises a key question – is overperception simply men and women perceiving different levels of interest, or as we believe, should that term be reserved for perceiving more interest than was originally communicated?
So, to answer our original question, would you know if the person was sexually interested? You, like our participants, would most likely be able to tell. Communication errors still do occur, of course, and further work is needed to better understand the particular types of contexts and people that are more likely to lead to sexual communication errors.