AI in VR identified cognitive load with 88% accuracy — News

AI in VR identified cognitive load with 88% accuracy — News-Medical

UA.NEWS22
October
2026
06:17

Researchers at the University of East Anglia found that artificial intelligence models can determine the level of cognitive load during virtual reality tasks by changes in pupil size and blink rate. As News-Medical reports, AI distinguished between low and high levels of mental workload in healthy volunteers with 88% accuracy.

How the study was conducted

The study involved 84 volunteers, including 42 people with chronic musculoskeletal pain, including arthritis and fibromyalgia. Participants performed increasingly difficult memory tasks in a virtual reality headset equipped with an eye-tracking system.

As the tasks became more difficult, participants’ pupils dilated and they blinked more often. The authors used AI models that analyzed these signals over time. Participants also reported very low levels of discomfort and motion sickness symptoms during VR sessions.

More current news is available on the UA.News Telegram channel Telegram.

The prospect of personalized therapy

The scientists believe that in the future, VR headsets will be able to track cognitive load and automatically modify the therapeutic scenario for a particular user. This could potentially help maintain a balance between task difficulty and a person’s comfort during non-drug therapy for chronic pain.

At the same time, the algorithms performed noticeably worse among people with chronic pain. The researchers suggest that future systems will require individual adjustment to each person’s eye responses. They also stressed the need for larger studies to determine whether adaptive VR therapy will improve real outcomes for patients. The work was published in the journal Scientific Reports.