News

This AI can identify your personality type by tracking eye movements

Researchers say it could in future be used to improve human-machine interactions.
Researchers say it could in future be used to improve human-machine interactions. Researchers say it could in future be used to improve human-machine interactions.

Scientists have developed new technology that uses artificial intelligence to track eye movements and determine someone’s personality type.

The team designed algorithms with machine-learning capabilities, ie, the ability for a computer to act without being explicitly programmed, using data gathered from 50 participants whose eye movements were recorded as they undertook everyday tasks using eye-tracking headsets.

They also assessed the participants’ personality traits using specially-designed questionnaires commonly used by psychologists.

The algorithms were trained to look at the Big Five personality traits – openness, conscientiousness, extraversion, agreeableness and neuroticism. The system was able to reliably recognise four of them (neuroticism, extraversion, agreeableness and conscientiousness).

The researchers say their work builds on previous studies in which it was observed that people with similar traits tend to move their eyes in similar ways.

Male robots.
Male robots. According to researchers, their findings could be used to improve human-robot interactions in the future (Lexx/Getty Images) (iLexx/Getty Images/iStockphoto)

According to Dr Tobias Loetscher, of the University of South Australia, their findings could be used to improve human-machine interactions in the future.

He said: “There’s certainly the potential for these findings to improve human-machine interactions.

“People are always looking for improved, personalised services. However, today’s robots and computers are not socially aware, so they cannot adapt to non-verbal cues.

“This research provides opportunities to develop robots and computers so that they can become more natural, and better at interpreting human social signals.”

The algorithms were developed by the University of South Australia in partnership with the University of Stuttgart, Flinders University and the Max Planck Institute for Informatics in Germany.

The findings are reported in the journal Frontiers in Human Neuroscience.