Skip to main content

Science Says, If You Like Instagram’s ‘Inkwell’ Filter, You Might Have Depression


Can you tell whether someone has depression just from looking at their social media accounts? According to scientists from Harvard University and the University of Vermont, there are some telltale patterns. Andrew Reece and Chris Danforth, two of the researchers involved in the study, found that the correlation between clinical depression and certain types of Instagram photos to be so strong that they managed to create an algorithm to detect it.

The study, described in more detail over at MIT Technology Review, only included 170 participants, 70 of whom had a diagnosis for clinical depression. The analysis drew from 40,000 photographs; in the case of the individuals with depression, the researchers only analyzed the photos that those users had taken before their diagnosis (and, therefore also, the photos they took before they might’ve received any sort of treatment for their depression).

Turns out, the phrase “feeling blue” makes a lot of sense, at least according to this data. The users who had depression tended to upload photos with a blue, black, or gray color palette. Also, these users tended towards using one filter in particular: “When depressed participants did employ filters, they most disproportionately favored the ‘Inkwell’ filter, which converts color photographs to black-­and-­white images.”

They also tended to have few likes on their photos, which makes sense, given that many of the symptoms of depression (e.g. fatigue, loss of interest in activities, and so on) can make it difficult to maintain a support network of friends, whether online or offline. Along those same lines, photos of faces that appeared on depressed users’ Instagram feeds often featured fewer people per photo, compared to other users’ average face count. This could mean that depressed people are uploading selfies, or simply that they have very few friends in their lives with whom to take photos.

Reece and Danforth hope that their findings can better help to detect potential signs of depression, even though the algorithm obviously doesn’t serve up an official diagnosis. Still, if you have a friend whose photos meet the description here, perhaps it’s worth dropping them a line?

(via Geek, image via Jens Karlsson/Flickr)

Want more stories like this? Become a subscriber and support the site!

The Mary Sue has a strict comment policy that forbids, but is not limited to, personal insults toward anyone, hate speech, and trolling.—

Follow The Mary Sue on Twitter, Facebook, Tumblr, Pinterest, & Google+.

Have a tip we should know? [email protected]

Filed Under:

Follow The Mary Sue:

Maddy Myers, journalist and arts critic, has written for the Boston Phoenix, Paste Magazine, MIT Technology Review, and tons more. She is a host on a videogame podcast called Isometric (, and she plays the keytar in a band called the Robot Knights (