AI & Machine Learning in Social Cognition: Building understanding of subjective notions of human attraction

July 19, 2021
a large group of people walk through an intersection while their faces are being recognized to demonstrate the use of ai in social cognition

Philosophers, authors, and poets have searched for a definition of beauty throughout history. Now, researchers from the University of Helsinki have used artificial intelligence (AI) to review electroencephalography (EEG) data to understand the subjective elements that make someone be classified as attractive.

The study, published in the February edition of the IEEE Transactions in Affective Computing Journal, used AI to interpret study participants’ brain signals to create artificial faces that appealed to what each participant found personally attractive.

How Machine Learning Understood Human Attraction

The study used a generative adversarial neural network (GAN) to generate hundreds of artificial portraits, which were then viewed by 30 volunteer participants.

The volunteers “were asked to pay attention to faces they found attractive while their brain responses were recorded via [EEG],” writes Aino Pekkarinen for Science Daily. The EEG recorded the participants’ immediate brain response to viewing each picture.

Using machine learning (ML), the researchers then analyzed the brain data, which was shared via a brain-computer interface to a generative neural network, known as generative brain-computer interfaces (GBCI). From this, new portraits were generated that were expected to be attractive to the participants based on their individual brain data.

“Testing them in a double-blind procedure against matched controls, they found that the new images matched the preferences of the subjects with an accuracy of over 80%,” writes Pekkarinen.

Michiel Spapé, one of the study authors, says this breakthrough is important because most studies in this field have been based on objective patterns. The Helsinki study shows that it’s possible to detect and generate images based on psychology and individual preference.

Future Implications & Uses of AI in Studies of Social Cognition

"If this is possible in something that is as personal and subjective as attractiveness, we may also be able to look into other cognitive functions such as perception and decision-making. Potentially, we might gear the device towards identifying stereotypes or implicit bias and better understand individual differences," says Spapé for Science Daily.

Having tools to detect implicit bias, an unconscious form of stereotyping, could have far-reaching benefits, though study authors caution there is much more work to be done, including increasing the diversity of the GAN’s source materials and a greater focus on cross-cultural efficacy.

“GBCI could provide qualitatively different insights with other subjective categorization tasks, such as recognizing images as trustworthy, benevolent, or powerful,” state the study authors in the implications and future work section. “What kind of person would the GBCI generate, but more importantly, what would this tell us about the individual? While this remains speculation, the results presented in the present study cause us to believe the GBCI could be a significant step forward in social cognition.”

Capitol Tech offers bachelor’s, master’s and doctorate degrees in computer science, artificial intelligence, and data science. Many courses are available both on campus and online. To learn more about Capitol Tech’s degree programs, contact