One highly desirable trait can dominate how you choose a romantic partner

The way people choose potential romantic partners may rely on the same mental decision-making system used for many other types of choices, according to a recent study. The research found that a computational model accurately predicted both who people selected as potential partners and how quickly they made those decisions. The study was published in Cognitive Science.

For decades, researchers have tried to understand what drives romantic attraction. While scientists have learned a great deal about which traits people prefer and which are dealbreakers, less is known about the mental processes that combine those preferences into a final decision. The current study sought to address that question by examining whether romantic partner choice can be explained by a broader theory of human decision-making known as Psychological Value Theory.

Rather than assuming that romantic decisions are fundamentally unique, the theory proposes that people evaluate potential partners using the same value-based mechanisms employed in many everyday judgments and choices. Led by Dale J. Cohen of the University of North Carolina Wilmington, the research team conducted two experiments using hypothetical romantic partner profiles.

In the first experiment, participants compared partners described by a single characteristic, allowing the researchers to determine how individual traits influenced partner value. In the second experiment, participants evaluated partners described by multiple characteristics simultaneously. This enabled the researchers to investigate how people combine several desirable and undesirable traits when making romantic decisions. Researchers measured both participants’ choices and the amount of time required to make each decision.

The findings provided strong support for Psychological Value Theory. Across both experiments, the model successfully predicted participants’ choices and reaction times, accounting for more than 85 percent of the variation in behavior. This level of accuracy suggests that the model captured many of the cognitive processes involved when people evaluate potential romantic partners.

Rather than simply identifying which partner participants chose, the model was also able to predict how quickly those decisions would be made. The second experiment revealed particularly important insights into how people evaluate multiple partner characteristics at once. Traditional theories often assume that individuals mentally add together all positive and negative qualities to calculate an overall value. However, the results suggested a different process.

As the authors explained, participants “integrate multiple features via a Biased Average algorithm, where the most positive feature holds disproportionate influence.” In other words, one highly desirable characteristic could have a greater impact on partner choice than several moderately desirable traits combined.

Cohen’s team concluded that “these findings indicate that initial romantic partner choice recruits a general-purpose, value-based decision mechanism, providing a computational framework that can be extended to model partner choice in more complex, real-world contexts.” The researchers argue that understanding these processes could help explain how people navigate the early stages of romantic relationships and why certain characteristics can have such a powerful impact on first impressions.

However, the study has important limitations. Participants evaluated hypothetical partner descriptions rather than interacting with real people. Real-world romantic decisions often involve emotional chemistry, social dynamics, and contextual factors that cannot be fully captured in laboratory experiments.

The study, “Psychological Value Theory: Predicting Initial Romantic Partner Choice From a General-Purpose, Computational Cognitive Model of Value-Based Choice,” was authored by Dale J. Cohen, Tyler D. White, and Shanhong Luo. It was published in 2026.