As it turns out, we’re not so good at hiding our realities on social media. Researchers from the University of Pennsylvania and Stony Brook University developed a new algorithm that can accurately predict future depression based on the types of words we use on Facebook.
After years of studying how the words people use reflect their inner feelings, researchers decided to test whether people diagnosed with depression had sent any linguistic “red flags” in the months prior to diagnosis. Indicators included mentions of hostility and loneliness; words like “feelings” and “tears”; and more first-person pronouns like “I” and “me.”
Nearly 1,200 people provided digital archives, of which 114 had a depression diagnosis. To build the algorithm, researchers looked at more than half a million Facebook updates from the years leading to diagnosis for those with depression and those acting as the control group. They learned that the red flags could predict future depression as early as three months before the first documentation of the illness in a medical record.
Source: Proceedings of the National Academy of Sciences