Eye Movements Offer Clues to Personality, AI Finds

Design Sem Nome 2025 12 17T163205.845 1
Computers are increasingly able to understand human behavior—and now they can do so by examining how people move their eyes. Researchers led by the Max Planck Institute for Informatics have created an AI-based software system that infers personality traits by analyzing eye movement patterns. Just as humans instinctively form impressions of others based on visual cues, this system uses eye behavior to estimate traits such as neuroticism, extroversion, and curiosity. The goal is to make interactions between humans and computers more social, efficient, and adaptable.
Image Credits: Look the computer in the eye: the new software analyses eye movements to draw conclusions about a person’s personality traits.

Computers are increasingly able to understand human behavior—and now they can do so by examining how people move their eyes. Researchers led by the Max Planck Institute for Informatics have created an AI-based software system that infers personality traits by analyzing eye movement patterns. Just as humans instinctively form impressions of others based on visual cues, this system uses eye behavior to estimate traits such as neuroticism, extroversion, and curiosity. The goal is to make interactions between humans and computers more social, efficient, and adaptable.

“Eye movements don’t just help us see the world; they also offer insight into our inner state. They reflect who we are, how we feel, and what we are doing,” explains Andreas Bulling, head of the Perceptual User Interfaces research group at the Max Planck Institute for Informatics and a member of the Cluster of Excellence at Saarland University. People naturally and unconsciously interpret social signals conveyed through the eyes. Bulling and his collaborators in Stuttgart and Australia are now teaching computers to do the same. This research could enable robots to recognize and respond to human nonverbal communication. To achieve this, the team developed a machine-learning system that analyzes eye-tracking data to predict an individual’s personality traits.

Personality Characteristics Commonly Used by Psychologists

The researchers applied artificial intelligence to train their software to predict four key personality traits commonly used by psychologists to describe individuals. Using this approach, the system was able to estimate how neurotic, agreeable, extroverted, and conscientious participants were, along with their level of curiosity. According to Andreas Bulling, the predictions are not yet precise enough for real-world use, but he expects the system’s accuracy to improve over time. His confidence is partly based on the potential to train the software with much larger datasets in the future, as well as additional information from a camera built into the eye tracker that records what the wearer is looking at.

The training and testing data came from 50 students at Flinders University in Australia, with an average age of 22. While the students spent about ten minutes walking around campus and purchasing coffee or other items from the campus shop, an eye tracker recorded their eye movements, focusing on where they looked and how frequently they blinked. After the experiment, the participants completed established questionnaires that psychologists have long used to assess personality.

Nonverbal Communication helps Robots Appear more Human

“The insights into nonverbal behavior gained from this study could, for instance, help robots behave in a more human-like way,” Bulling explains. Such systems could interact with people more naturally, which would allow them to operate more flexibly and effectively. The software could also support individuals with autism or other conditions that make nonverbal communication challenging, by helping them interpret others’ visual cues and manage their own eye movements—for example, learning how to avoid prolonged staring.

At the same time, Bulling acknowledges that the technology raises concerns about computer-assisted personality assessment, which could be misused by corporations or authoritarian governments already engaged in digital behavior analysis. He stresses, however, that the system is currently far from capable of reliably identifying personality traits without human involvement, especially since it still requires users to wear an eye tracker positioned directly in front of the eye. Even if assessing personality were to become easier in the future, Bulling notes that, like many technologies, it could be applied for both positive and negative purposes. “As scientists, our role is to develop the technology and demonstrate what is possible,” he says. “How it is ultimately used must be governed by society and the law.”


Read the original article on: SciTechDaily

Read more: New AI Technology Astonishingly Excels at ‘Understanding’ Human Thoughts

Scroll to Top