You are where you look: AI identifies readers through eye movements

German researchers have developed a system that learns to associate patterns in eye movement with individuals.

Writing in arXiv, the authors highlight that eye movement patterns vary significantly between individuals.

“This property makes them interesting for biometrics,” they emphasised.

The researchers suggest that tracking the eye movements of an individual while they read has several potential applications.

“Users can be identified unobtrusively while having access to a document that they would read anyway, saving time and attention,” the authors observed.

In a test involving 62 readers, the artificial intelligence system was able to identify readers with an accuracy of 91.53%.

Research published in April this year investigated the link between gaze patterns and personality. 

The study, which was published in Frontiers in Human Neuroscience, used eye movements to predict whether study participants were neurotic, extroverted, agreeable or conscientious.