What we think we like in a romantic partner – and what we prefer – don’t always match up

If I like humor, it could be that I’ve gone on dates with funny people, and I liked them.


People have ideas about the attributes that they like in their partners. These ideas about liking are called summarized attribute preferences. But where do summarized preferences come from? And, more importantly, do they reflect our actual experiences?

A new study form the University of Toronto examined how people form summarized attribute preferences and whether they predict situation selection. Scientists found that what people think they like in a romantic partner and what they actually like can often be two different things.

The scientists discovered a weak relationship between what people believe they enjoy and what motivates their like. In actuality, people’s conceptions of like and actual experiences with liking can end up forecasting various choices in behavior.

This effect has been tested across four studies involving more than 1,300 participants. In the first three studies, participants’ ideas about how much they liked a trait in a potential romantic partner were barely correlated to how much they liked it.

Minor changes in the environment can also have an impact on people’s perceptions of how much they loved a particular trait. In the previous study, participants were asked to rate how much they valued traits like confidence. The participants then indicated their interest in signing up for various dating websites based on how much they liked a selection of online dating profiles.

The findings demonstrated that different decisions were predicted by what participants believed they liked and what they liked. For instance, their interest in signing up for a free dating service trial with pictures of confident people did not correspond to their perceptions of how much they enjoyed confidence. Participants’ propensity to jump into the dating pool was predicted by how much they liked confidence after having it.

Andre Wang, an assistant professor in the department of psychology at U of T Scarborough said“After the free trial, ideas about liking didn’t matter anymore. At that point, what matters more are experiences of liking. Once you experience something, that becomes your guide.”

“Ultimately, people’s ideas about what they like, although useful in many situations, are no substitute for actual experiences. Understanding the distinction between what we think we like versus what drives us to like something can be useful in various situations. For example, it could help people predict where to live, what to buy and what they prefer in a romantic partner.”

“It’s possible people unnecessarily rule out potential partners based on certain traits they think they like but have never actually experienced in person.”

“It could be that people are so constrained by their ideas about liking that they are limiting their dating pool.”

“They could be filtering out people in advance who might make them truly happy.”

Journal Reference:

  1. da Silva Frost, A., Wang, Y. A., Eastwick, P. W., & Ledgerwood, A. (2022). Summarized attribute preferences have unique antecedents and consequences. Journal of Experimental Psychology: General. DOI: 10.1037/xge0001242
Latest Updates