Eye movements linked personality
The eyes really are a window to the soul, according to scientists who have created a 'mind-reading' AI that can predict your personality from looking at small eye movements and blinking.
You[r] eyes may be able to reveal more about you than you realise.

Scientists have created a 'mind-reading' AI that can predict your personality from looking at pupil movements and blinking.

Curious people tend to look around more and open-minded people stare at abstract images for longer periods of time, researchers revealed.

Scientists, led by Tobias Loetscher from the University of South Australia, used machine learning to understand how eye movements and personality are related.

Forty-two students wore eye-tracking smart glasses while walking around campus, writes New Scientist.

The students also filled out questionnaires that rated their personalities.

This questionnaire broke down personality into the 'Big Five' traits used widely in psychology; extraversion, neuroticism, conscientiousness, agreeableness, and openness to experience.

Personality traits characterise an individual's patterns of behaviour, thinking, and feeling', researchers wrote in their paper published in Frontiers in Human Neuroscience.

'Studies reporting relationships between personality traits and eye movements suggest that people with similar traits tend to move their eyes in similar ways.'

Researchers found that people who were neurotic usually blinked faster while people who were open to new experiences moved their eyes more from side-to-side.

People who had high levels of conscientiousness had greater fluctuations in their pupil size.

Optimists spent less time looking at negative emotional stimuli (such as image of skin cancer) than people who were pessimistic.
AI predicts personality traits
'Besides allowing us to perceive our surroundings, eye movements are also a window into our mind and a rich source of information on who we are, how we feel, and what we do', researchers wrote.

'The proposed machine learning approach was particularly successful in predicting levels of agreeableness, conscientiousness, extraversion, and perceptual curiosity'.

Scientists found the machine is currently between seven and 15 per cent better than random chance at predicting these traits.

However, it is no better than random chance at predicting openness.

Researcher do not know why there are these links but say that it will help them to teach robots to be more socially aware.

It could be put in smartphones that understand and predict our behaviour, potentially offering personalised support.

They could also be used by robot companions for older people, or in self-driving cars and interactive video games.

Researchers warn that the technology would have to be regulated so it was not misused by marketers.

'Improving automatic recognition and interpretation of human social signals is an important endeavor, enabling innovative design of human-computer systems capable of sensing spontaneous natural user behavior to facilitate efficient interaction and personalization', researchers wrote.