A new artificial intelligence platform can decode pain intensity directly from brainwaves, using EEG signals to provide an objective, real-time measure of physical suffering — without relying on a patient's ability to describe their pain.
The Research
Led by Principal Researcher An Jinung at DGIST and Professor Jeon Seong-chan at GIST in Korea, the study involved 41 participants who received thermal stimuli while their EEG was recorded. Instead of training an AI on subjective pain scores (like the Visual Analogue Scale), the team developed a dual-model, self-correcting algorithm. Two AI models cross-compared their predictions, learning only from highly reliable data points. This effectively filtered out individual biases in pain expression.
The platform isolated delta wave activity specifically at the F7 and F8 frontal nodes — the left and right anterior temporal lobes — proving that this brainwave signature tracks directly with physical pain intensity. The model maintained stable, accurate predictions even when exposed to new stimulus environments not seen during training, outperforming traditional neural networks. The work was funded by the National Research Foundation of Korea.
Why It Matters
For anyone curious about cognition, this research shows that brain activity patterns — in this case delta waves — can serve as a biological ruler for subjective experiences like pain. The same approach could eventually be adapted to decode other internal states, such as mental fatigue, attention, or emotional distress. It also highlights how AI, when carefully designed, can overcome human reporting biases and unlock new windows into the brain's inner world.
What You Can Do
While you can't measure your delta waves at home, you can practice becoming more aware of your internal states. Try a simple body scan: close your eyes for 30 seconds, and note any physical sensations without judging them. This mindfulness exercise strengthens your ability to observe your own cognition, a skill that complements any brain training regimen.
Source: Neuroscience News
Curious about your own brain? Take our free adaptive IQ test or try 306 brain training levels.
