For people living with chronic pain, relief could one day come from a virtual world that responds to their needs simply by watching their eyes. Researchers at the University of East Anglia (UEA) have discovered that subtle changes in pupil size and blinking patterns could help artificial intelligence (AI) tailor virtual reality (VR) treatments to individual patients.
Chronic pain affects an estimated 1.5 billion people worldwide and remains a leading cause of disability. While VR is increasingly being explored as a drug-free approach to pain management, most existing programmes provide the same experience to everyone, regardless of their changing pain levels or mental demands.
VR therapy can transport patients into peaceful forests, guide them through relaxation exercises or engage them in interactive games and puzzles. These immersive experiences may help divert attention from pain, reduce stress and anxiety, and potentially change how the brain processes pain over time.
“Virtual reality has enormous therapeutic potential,” said lead researcher Dr Jordan Tsigarides from UEA’s School of Health Sciences. However, he explained that conventional VR programmes rarely account for fluctuations in patients’ mental effort. By monitoring eye movements and responses, future systems could automatically adjust activities to prevent mental overload while maintaining therapeutic engagement.
The study involved 84 volunteers, including 42 people living with chronic musculoskeletal conditions such as arthritis and fibromyalgia. Participants completed increasingly difficult memory challenges while wearing VR headsets equipped with eye-tracking technology. Researchers monitored changes in pupil size and blinking frequency as the mental demands increased.
The findings revealed that pupils widened and blinking became more frequent during challenging tasks. AI models analysing these signals distinguished between low and high mental workload with up to 88 per cent accuracy among healthy volunteers. Participants also reported minimal motion sickness or discomfort, suggesting the technology was generally well tolerated.
The discovery raises the possibility of developing intelligent VR treatments that continuously monitor users and adjust their experiences in real time. A headset might simplify an activity when someone becomes mentally overwhelmed or increase engagement when additional stimulation is appropriate, potentially making therapy more comfortable and effective.
However, an important challenge remains. Although the AI performed well among healthy participants, its accuracy was considerably lower among those experiencing chronic pain. Researchers believe future systems may require individual calibration to recognise how each patient’s eyes respond to mental effort, rather than relying on a standardised approach.
Dr Tsigarides said the research could eventually extend beyond chronic pain to conditions such as anxiety and depression, as well as education and professional training. Nevertheless, larger studies are needed to establish whether adaptive VR actually improves patient outcomes. For now, the findings offer a promising glimpse of a future in which pain treatment becomes more personalised, responsive and potentially drug-free.
More information: Jordan Tsigarides et al, VR-FOCUS: evaluating eye tracking during a virtual reality N-back task as a marker of cognitive load in chronic pain, Scientific Reports. DOI: 10.1038/s41598-026-74975-6
Journal information: Scientific Reports Provided by University of East Anglia