What Your Online Photos Say about Your Mental State
We convey our moods with body language and behavior, and as it turns out, with the images we post on social media. New research suggests computers can accurately detect depression from clues in Instagram photos at a rate that is 70 percent more reliable than the 42 percent success rate of general practitioners. The study was led by researchers at the University of Vermont in Burlington and Harvard University in Cambridge, Massachusetts, and was published in EPJ Data Science.
Researchers recruited 166 volunteers to share their Instagram feeds – a total of 43,950 photos -- and their mental health history. About half of the study participants reported having been diagnosed with clinical depression within the past three years. Researchers then analyzed the photos for brightness, color, and shading. They found that people who had not been diagnosed with depression chose filters that gave photos a warmer, brighter tone, while people with depression chose black-and-white filters more often.
They also determined that faces in photos may provide additional depression signals. People with depression are more likely to post photos that include faces, but their photos generally contain fewer faces overall than those of people without depression. According to researchers, this may indicate a tendency to interact in smaller social settings or take more selfies.