Depression is often called “the blues,” and two scientists think they’ve found the reason. They created a computer algorithm that can detect depression by looking at someone’s Instagram, MIT Technology Review reported yesterday. This was possible because depressed people’s pictures contain similar colors, filters and content.
Andrew Reece of Harvard University and Chris Danforth of the University of Vermont surveyed 170 Instagram users about their mental health. 70 were clinically depressed. They studied the 100 most recent posts of the non-depressed subjects. For depressed subjects, they studied 100 posts made before their diagnosis.
Reece and Danforth studied color, vividness, the number of faces in a photo, likes and comments. People with depression post pictures with fewer people and got fewer likes. This may be because because they’re less likely to socialize frequently. Depressed Instagram users also posted darker, bluer and grayer pictures. The preferred filter of healthy subjects was the lightening “Valencia.”
“When depressed participants did employ filters, they most disproportionately favored the ‘Inkwell’ filter, which converts color photographs to black-and-white images,” Reece and Danforth said.
Reece and Danforth created an algorithm to identify these signs. It found the depressed subjects based on their Instagrams 70 percent of the time. That’s a higher rate of detection than primary care doctors. The algorithm’s success gives the researchers hope it can be used for early detection.
“These findings support the notion that major changes in individual psychology are transmitted in social-media use, and can be identified via computational methods,” Reece and Danforth said.
Lead image credit: Jean Gerber on Unsplash.