They say that beauty is in the eye of the beholder, but in reality it goes much deeper than that.
The concept of physical beauty is in the mind, determined by the features that we find attractive in other people’s faces. These subtle preferences represent some of our most personal inner thoughts – but that doesn’t mean they can’t be controlled and maybe even predicted.
In a new study, researchers used electroencephalography (EEG) measurements to determine what kind of facial features people found attractive, and then passed the results to an artificial intelligence (AI) program.
The machine learning system – called a generative adversarial neural network (GAN) – was first able to familiarize yourself with the types of faces that individual people found desirable, and then fabricate completely new faces specifically designed to please: on bespoke visions of synthesized beauty, as unattainable because they were perfect.
The experiment, led by a team of psychologists and computer scientists at the University of Helsinki in Finland, was like a massive Tinder session for the 30 volunteers who took part.
Except for a few major differences.
While the participants sat in front of a computer screen and showed them a series of faces, none of the faces shown were real people, but realistic-looking artificial portraits generated from a dataset of approximately 200,000 celebrity images.
Unlike normal use of Tinder, the participants also wore elastic caps with electrodes designed to measure their brain activity while looking at the faces. They also didn’t have to swipe right when they saw someone they liked – that was taken care of.
“They didn’t have to do anything but look at the images,” explains cognitive neuroscientist Michiel Spapé. “We measured their immediate brain response to the images.”
Those individual measures of neural activity were then assessed by the GAN, which was able to interpret the brain responses in terms of how attractive each artificial face was perceived by the viewer.
Using that data, the GAN was then able to generate new faces based on people’s EEG attraction IDs.
In a second experiment, these newly invented faces were then shown back to the volunteers, who rated them in terms of attractiveness, alongside other images of randomly generated faces.
Ultimately, the results validated the researchers’ test, with participants rating the custom images as attractive in about 80 percent of the cases, while the other faces were selected only 20 percent of the time.
While this is only a small study, it is just another example of how sophisticated AI systems are increasingly understanding what drives us – even in intimate and often unspoken concepts, such as the domain of personal attraction.
“Assessing attractiveness is especially important, as this is such a poignant psychological property of the stimuli,” says Spapé.
“If this is possible in something as personal and subjective as attraction, we may also be able to look at other cognitive functions, such as perception and decision-making. We may focus the device on identifying stereotypes or implicit biases and better understand individual differences.” “
The findings are reported in IEEE transactions on affective computers