By Alimat Aliyeva
Scientists at Ben Gurion University in Israel have developed a machine learning model that predicts romantic attraction emotion with moderate accuracy in conditions that simulate the behavior of dating apps. azernews
Report.
This study was published in the Journal of Biology and Medicine (CBM).
Sixty-one students aged 23 to 32 years old participated in the experiment. While looking at the photos of potential partners, their brain activity was recorded using electroencephalography (EEG). Participants were told they had received “feedback” by indicating who they found attractive. This setup successfully replicated the core mechanism of the dating app: attraction, followed by a possible rejection.
Machine learning algorithms analyzed the electrical responses of the brain, known as induced potentials. The model predicted participants' attractive responses with an accuracy of 71.3% and rejected rejection responses with an accuracy of 81.3%. Interestingly, the researchers found that the algorithm performed even better in “noisy” participants (the “noisy” participants due to a clearer and more consistent neural patterns for less attractive people), and according to the researchers.
“Analyzing the EEG signal allows you to predict user decisions for dating apps, even if you swipe right or left for example,” the author explained. “This approach provides deeper insight into emotional responses and reveals whether a person is attracted to them or experiences negative emotions associated with rejection.”
This breakthrough opens the door to a more personalized, emotionally aware dating platform, potentially transforming the way users connect and interact in digital dating situations. It also raises interesting ethical questions about privacy and the use of neural states in everyday technology.
