THE LOOK holds a thousand secrets and many could be unveiled by artificial intelligence . An international team of researchers from Australian and German institutions has sought to better understand the link between personality and eye movements by developing an algorithm of machine learning. It is a type of computer code that learns without the need to be programmed in a specific way and is able to identify the personality traits of each analyzing the movement of the eyes. A discovery that could help computers understand and interact better with human beings.
For the study , published in the journal Frontiers in Human Neuroscience, the researchers have provided 42 students from Flinders University, in the south of Australia, with special eyeglasses for eye tracking. They then asked the participants to go around the campus and enter a shop to shop, then fill out a questionnaire, which assessed them on five personality traits: openness, conscientiousness, extroversion, joviality and neurosis. Using a state-of-the-art machine learning method and a rich set of features that encode different eye movements, the researchers were able to reliably predict four of the character traits examined - neurosis, extroversion, joviality and conscientiousness - as well as curiosity perceptive, based only on eye tracking data.
Psychologists have long maintained that the personality influences our visual approach to the world. For example, those who are curious tend to look around and those with an open mind linger longer on abstract images. "Several previous studies have suggested that the way we move our eyes is modulated by who we are, by our personality," he told Digital Trends Andreas Bulling , among the authors of the research, the German Institute for computer science professor Max Planck . "For example, studies that report relationships between personality traits and eye movements tell us that people with similar traits tend to move their eyes the same way." So the machines can learn to distinguish and understand our state of mind by looking into our eyes.
The results of the study could allow, in other words, the design of computer systems, including robots, smartphones and self-driving cars, able to automatically read user personalities on the basis of this data and customize, as a consequence, the human-machine interaction