Scientists from Ben-Gurion University in Israel have developed a machine learning model that predicts feelings of romantic attraction with moderately high accuracy in conditions simulating the operation of a dating app, Azernews reports.

The study was published in the journal Computers in Biology and Medicine (CBM).

Sixty-one students aged between 23 and 32 participated in the experiment. While viewing photos of potential partners, their brain activity was recorded using electroencephalography (EEG). Participants indicated who they found attractive and then received "feedback"-purportedly an assessment from those they liked, but in reality, it was simulated. This setup successfully replicated a core mechanism of dating apps: attraction followed by possible rejection.

Machine learning algorithms analyzed the brain's electrical responses-known as evoked potentials. The models predicted participants' attraction reactions with an accuracy of 71.3%, and rejection reactions with an accuracy of 81.3%. Interestingly, the algorithms performed even better with "picky" participants-those who marked fewer people as attractive-likely due to more distinct and consistent neural patterns, according to the researchers.

“By analyzing EEG signals, we can predict user decisions in dating apps-for example, whether they swipe right or left,” the authors explained.“This approach provides deeper insight into emotional responses, revealing whether a person feels attraction or experiences negative emotions associated with rejection.”

This breakthrough opens the door to more personalized and emotionally aware dating platforms, potentially transforming how users connect and interact in the digital dating landscape. It also raises intriguing ethical questions about privacy and the use of neurodata in everyday technology.

MENAFN04102025000195011045ID1110148800