The body's pain response reflected in tiny facial movements
For generations, medicine has asked patients to translate their suffering into numbers, trusting language to carry what the body already knows. Researchers at Rutgers University have found that pain writes itself across the face in movements too small for the human eye to catch — micromovements that mirror the heart's own distress signals, offering a physiological truth that requires no words. The work, published in Frontiers in Neuroscience, holds particular promise for those who cannot speak their pain: children, stroke survivors, people living with dementia. It is, in essence, an attempt to let the body testify on its own behalf.
- The familiar 1-to-10 pain scale has long masked a quiet crisis — for millions of non-verbal patients, there has simply been no reliable way to measure suffering at all.
- A Rutgers study of 45 adults found that as pain intensified, heart rhythms grew irregular and the face responded within seconds, most visibly around the eyes — a connection so direct it could be tracked by AI-assisted video analysis.
- An unexpected finding complicated the picture: heavy cognitive tasks weakened the face-heart pain signal, suggesting the brain can partially crowd out pain — a clue that mental engagement might itself be a therapeutic tool.
- The technology still requires pairing facial video with heart monitors, and the study was modest in scale, leaving larger and more diverse populations yet to be tested.
- A smartphone application is already in development through Rutgers spinoff Neuroinversa LLC, pointing toward a future where a brief facial scan replaces the emoji pain chart hanging on the clinic wall.
For decades, the ritual has been the same: a doctor asks where pain falls on a scale of one to ten, and a patient reaches for words to describe something the body is already expressing on its own. Elizabeth Torres, a psychology professor at Rutgers University-New Brunswick, has spent years asking whether that translation step could be bypassed entirely.
Working with doctoral researcher Mona Elsayed, Torres published findings in Frontiers in Neuroscience showing that pain leaves a physiological signature in the face — rapid, subtle micromovements invisible to the naked eye that correspond directly to changes in heart rate variability. In a controlled study of 45 adults experiencing brief pressure pain, AI-assisted video analysis revealed that as discomfort intensified, cardiac rhythms grew irregular and those disturbances surfaced almost immediately in facial muscle activity, especially around the eyes.
The research surfaced a secondary insight: when participants performed cognitively demanding tasks, the link between facial signals and heart rhythm weakened. Pain, it appeared, could be partially displaced by mental engagement — a finding with potential therapeutic implications for redirecting attention away from discomfort.
Torres's Sensory Motor Integration Lab has long used mathematical modeling to decode internal states through subtle movement in people with autism, Parkinson's, and other neurological conditions. That foundation now points toward a population with urgent need: non-verbal patients — young children, stroke survivors, people with dementia — for whom caregiver interpretation has been the only available tool. This method offers something more direct: access to the physiology itself.
The practical path forward involves shrinking the technology. Current methods require specialized heart monitors alongside facial video, but Torres envisions smartphones eventually capturing the same data. A digital dashboard tracking pain day to day — showing whether medications are working, how quickly they take effect — could replace the emoji-based charts common in clinical settings today. Neuroinversa LLC, a Rutgers spinoff, is already developing a smartphone application based on the licensed technology. Larger studies across more diverse populations, particularly those with chronic pain, will determine whether the approach holds at scale — and whether medicine can learn, at last, to listen to what the body has always been saying.
For decades, doctors have asked the same question in the same way: On a scale of one to ten, how much does it hurt? It's a ritual so familiar that few stop to question its limits. But Elizabeth Torres, a psychology professor at Rutgers University-New Brunswick, has been asking a different question: What if the body itself could tell us the answer, without waiting for words?
Torres and her doctoral researcher Mona Elsayed published findings in Frontiers in Neuroscience suggesting that pain leaves traces invisible to the naked eye—tiny, rapid movements across the face that correspond directly to what someone is experiencing physiologically. These facial micromovements, too subtle for human observation to catch, spike in response to discomfort and align with measurable changes in heart rhythm. The discovery emerged from a controlled study of 45 adults who experienced brief episodes of pressure pain while researchers recorded their facial expressions and monitored their heart rate variability using video analysis and artificial intelligence.
The results were striking. As pain intensified, the heart's rhythm became increasingly irregular, and those disturbances showed up almost immediately in the face—most noticeably around the eyes. The connection was so direct that within seconds, the researchers could see the body's pain response reflected in facial muscle activity. "Every individual has a different threshold for pain tolerance," Torres explained. "By measuring that response directly from the body's own signals, we can begin to tailor care in a much more individualized way." The motivation was straightforward: move beyond a one-size-fits-all approach to something grounded in each person's actual physiology.
The research also revealed something unexpected about how the brain processes pain. When participants engaged in tasks requiring significant cognitive effort—memory work or sustained attention—the connection between facial movements and heart rhythm weakened. Pain, it seemed, could be crowded out by mental engagement. Tasks involving touch and physical manipulation, by contrast, showed the strongest link between facial signals and physiological distress. This finding hints at a potential therapeutic angle: cognitive engagement might serve as a natural distractor, offering clinicians a tool for redirecting attention away from discomfort.
Torres's work builds on years of research in her Sensory Motor Integration Lab, where she has used mathematical modeling to decode internal states through subtle body language in people with autism, Parkinson's disease, and other neurological conditions. That foundation proved crucial here. The same approach that revealed physical distress in nonverbal patients could now offer clinicians an objective window into pain for populations who cannot describe their symptoms: young children, stroke survivors, people with dementia. Right now, caregivers interpret what they observe, and while valuable, those interpretations remain incomplete. This method offers something different—direct access to the physiology itself.
The practical implications are significant. Current assessment requires pairing facial video with specialized heart monitors, but Torres believes widely available tools like smartphones could eventually capture the same data. Advances in video analysis and artificial intelligence have made it possible to detect physical markers that once demanded specialized equipment. A short facial scan could replace the emoji-based pain charts that currently hang in many clinical settings. Instead, patients and clinicians could access a digital dashboard showing pain levels tracked day to day, offering a much more precise way to monitor whether medications are working, how quickly they take effect, and whether adjustments are needed.
Torres and collaborators are already translating the technology into a smartphone application through Neuroinversa LLC, a Rutgers spinoff company that licensed the technology. The app remains in development, but the vision is clear: pain measurement that gives people a sense of control over their own biorhythms. The research is still early. The study was modest in size, though statistically powerful given the sensitivity of the personalized micromovement metrics. The next phase will test the approach in larger, more diverse populations, particularly those living with chronic pain. If it works at scale, it could reshape how clinicians assess suffering—not by asking, but by listening to what the body already knows.
Notable Quotes
Every individual has a different threshold for pain tolerance. By measuring that response directly from the body's own signals, we can begin to tailor care in a much more individualized way.— Elizabeth Torres, psychology professor, Rutgers University-New Brunswick
A higher cognitive load essentially crowds out the pain. This kind of engagement may act as a natural distractor, offering a potential therapeutic tool for redirecting attention.— Elizabeth Torres
The Hearth Conversation Another angle on the story
Why does it matter that we can see pain in the face if we can already ask someone how much it hurts?
Because not everyone can answer. A child with autism, a stroke survivor who's lost speech, someone with advanced dementia—they can't tell you. Right now we guess based on behavior. This gives us something objective.
But couldn't someone just hide their pain? Control their facial expressions?
That's the beauty of it. These aren't conscious movements. They're micromovements—so fast and subtle that you can't fake them or suppress them. The body's physiological response shows up whether you want it to or not.
So it's like a lie detector for pain?
Not quite. A lie detector measures stress and arousal. This measures something more specific—the actual dysregulation of your nervous system in response to physical discomfort. It's the difference between detecting that something's wrong and knowing what's actually happening.
The study mentions that cognitive tasks weakened the pain signal. Does that mean pain isn't real if you're distracted?
No. It means the brain can modulate how much attention it pays to pain. The pain is still there physiologically, but when you're deeply engaged in something else, your nervous system doesn't respond as dramatically. That's useful therapeutically—it suggests distraction isn't just psychological comfort, it's a real neurological mechanism.
What happens when this becomes a smartphone app? Won't people obsess over their pain numbers?
Possibly. But right now they have almost no objective data about their own pain. A dashboard at least gives them information. Whether they use it wisely—that's a different question, and it's one clinicians will have to navigate.