Can adults be trained to better recognize and avoid sick faces?

Previous research from the University of Miami showed that children can recognize sick faces, demonstrating that even kids can pick up on illness cues, though accuracy varies by age. Now, researchers have taken this line of research one step further to answer the question: Can this skill be improved through training?

Building on their previous work, researchers from the University of Miami, James Madison University, and Duke University have published new findings showing that adults can indeed be trained to avoid sick people more effectively using facial cues alone. The study, published in the journal Vision, is the first to test whether sick face perception—a crucial component of the behavioral immune system—can be enhanced through experience.

“We started this research slightly before the COVID-19 pandemic, so it was interesting timing,” said Tiffany Leung, the lead author and a developmental psychology doctoral student at the University of Miami College of Arts and Sciences. “Our general goal has always been to improve public health—to reduce how frequently contagious illnesses are being passed around among individuals.”

Training the eye to spot sickness

The study involved 133 adults, mainly University of Miami undergraduates, who participated in a carefully designed experiment. Participants viewed 32 photos of 16 people—each photographed while healthy and while sick with a contagious illness like COVID-19, flu, strep throat, or a cold. This unique approach allowed researchers to control for individual differences such as age, attractiveness, and facial structure.

Half of the participants received brief disease training consisting of an interactive story, an educational video about infectious diseases, and practice trials with feedback on identifying sick faces. The training emphasized key facial features of illness, particularly drooping eyelids and relaxed facial muscles around the mouth.

“One of the things we’ve seen with digitally altered versions of pictures is that they’re not as realistic,” explained Leung. “What was important to us was getting this more real-world representation of what it is like to be sick, including the variability.”

Improved avoidance

While both groups started at about 65% accuracy in avoiding sick faces, the disease-trained group improved to 70%, but the control group showed no improvement. Remarkably, the training didn’t improve explicit recognition of sick faces—both groups performed similarly when directly asked to identify who was sick. Instead, the training enhanced participants’ unconscious avoidance behaviors.

Using eye-tracking technology, researchers discovered that training changed how people looked at faces. The control group spent more time staring at sick faces, while the trained group looked more evenly between sick and healthy faces, making systematic comparisons. Trained participants also made more back-and-forth glances between the mouth regions of faces, consistent with detecting the drooping mouth corners characteristic of illness.

“The disease training group was far more likely to mention the eyes (92% versus 66%) and mouth (40% versus 18%) as features they used” to identify sick faces, Leung noted.

Real-world applications

The findings have significant implications for public health.

“I think there are implications for being able to develop programs for classrooms to teach children how to avoid disease and also in health care settings by training professionals to recognize subtle cues early on,” said Leung.

Elizabeth Simpson, an associate professor of psychology and an author on the study, emphasized the potential applications of the research findings for children. “While this improvement in adults was small, we’re predicting a larger improvement in children, who have a harder time recognizing sick faces and therefore may benefit more from this support,” she said.

Addressing ethical concerns

The research team acknowledged important ethical considerations around teaching people to identify the sick. In the Department of Psychology, some faculty members work with individuals who have obsessive-compulsive disorder and anxieties surrounding illness, particularly after the COVID-19 pandemic.

“It’s important that we emphasize that we’re not teaching people to be hypervigilant about disease, but rather we’re trying to improve accuracy,” said Leung. “We want to be able to tell when a sick person is sick, and when a healthy person is healthy.”

Looking ahead

The research team continues to expand its disease detection research. The team is currently recruiting children ages 5–9 to test whether similar training can improve sick face perception in young people. The researchers are also exploring individual differences, with preliminary findings suggesting women may have better sick face recognition skills than men.

“We have lots of individual differences that we’re interested in,” said Simpson. “We’re looking at whether, during pregnancy—when the immune system is suppressed and women are more vulnerable—they also show enhanced sick-face recognition.”

The team is also expanding globally, collaborating with researchers in China to test the generalizability of their findings across cultures. This collaboration is with Guangyu Zeng at the University of Hong Kong, Shenzhen, who is one of Simpson’s former Ph.D. students.

The researchers emphasized that their work wouldn’t be possible without volunteers willing to share photos of themselves while sick and healthy. For more information about how you or your child can contribute to this research, visit the Social Cognition Lab. To learn more about this topic, children can also visit the researchers’ online article for kids.