Your Instagram knows if you're depressed or not
Think you're just sharing snaps of your morning coffee, yoga routine or adorable fur babies on Instagram? According to a new study, you may be inadvertently sharing something a lot more personal than that: your mental health status.
Depression, anxiety and other mental illnesses are incredibly common in our society — 1 in 4 women will experience mental illness at some point in her life — yet it can still be hard to talk about. Even if we're not actually saying (or typing) about these feelings, depression and anxiety still manifest in our posts on social media according to research pre-published on Arvix.
Researchers recruited 166 Instagram users (who had a total of 144,000 pictures uploaded) and divided them into two groups: those who had been diagnosed with a mental illness and those who hadn't. They then used a special computer program to analyze the pictures of each person. But instead of looking at the content, like a human would, the program examined the photo's color, composition, saturation, number of people and type of Instagram filter used. A group of unrelated people was then asked to analyze the same pictures, looking for signs of sadness or depression.
What they found may surprise you: Not only could the computer predict which people had depression or anxiety, but it was better at it than the humans were! (According to a separate study, we may not even know when we're depressed!)
The computer was able to be so accurate because it found patterns that were consistent in how mentally ill people share their experiences of the world. For instance, depressed Instagrammers posted more pictures that were blue, gray, or dark-hued and were very partial to the Inkwell filter while non-depressed users had a lot more color and loved the Valencia filter.
But it's not just the picture that says a thousand words, it's also how people react to it. The researchers found that the more comments a picture got, the more likely the poster was depressed. However, the more likes the picture got, the less likely the poster had depression. (Scrambling to check your likes vs. comments on Instagram now? And then on all your friends accounts? Same.)
One thing the researchers didn't mention, though, was the interplay between social media and mental illness. Previous research has found that the more you use social media, the more at risk you are for depression, low self-esteem, anxiety and other mental health problems. So perhaps depressed people post differently on social media, but perhaps social media is also making them more depressed?
The point of their Insta-research, however, isn't to diagnose or even treat depression, said the study authors Andrew Reece of Harvard, and Christopher Danforth of the University of Vermont. "We foresee a more mature version of this tool being used more in the context of screening and assessment, rather than treatment," they told the Daily Dot. "The algorithm we used looks for complex, systemic patterns across many data points to infer clues about individual psychology. If someone posts a dark, bluish photo to Instagram, it shouldn't necessarily be a red flag — that person could just like photos of whales, or blueberries."
In the meantime, instead of wondering how a computer can get so good at understanding our friends' inner workings, maybe we should spend more time thinking about how we can better understand and support our friends with mental illness — in the real world.