Instagram can tell if you are depressed

<i></i>
 Photo: Shutterstock

If you've noticed a friend's Instagram posts have been bluer or darker of late, it might be worth asking them how they're feeling. A new study suggests it could be a clue that they're suffering from depression.

"Photos posted to Instagram by depressed individuals were more likely to be bluer, greyer, and darker, and receive fewer likes," the researchers from The University of Vermont and Harvard University, write in a paper published this week. "Depressed Instagram users in our sample had a preference for filtering out all colour from posted photos, and showed an aversion to artificially lightening photos, compared to non-depressed controls."

As part of the study, published in the journal EPJ Data Science, the research team recruited 166 people and asked them to share their Instagram photos - 43,950 in total. Approximately half of the particpants had been diagnosed with clinical depression over the past three years. After analysing the pictures, the team found that for those diagnosed with depression, the filter "Inkwell," which renders pictures black and white, was the most popular, (as well as choosing no filter at all) while "healthy" individuals most commonly chose brighter options, such as Valencia. 

"In other words, people suffering from depression were more likely to favour a filter that literally drained all the colour out the images they wanted to share," the authors note.

Instagram photos posted by depressed individuals had were bluer, (higher Hue) grayer (lower Saturation) and darker (lower Brightness) compared with photos posted by healthy individuals. Image/ EPJ Data Science.

Along with darker filters, there was another key difference between the Instagram feeds of the two groups of participants. Depressed users were more likely to post photos with people's faces - but the photos had fewer faces on average than healthy individuals' snaps. "Fewer faces may be an oblique indicator that depressed users interact in smaller settings," the authors explain, adding that those suffering from depression may also take fewer selfies.

However, "This 'sad-selfie' hypothesis remains untested," they caution. The finding, the researchers continue, is also consistent with research demonstrating that "reduced social interactivity is an indicator of depression".

In the second part of the study, the researchers recruited further volunteers to see whether they could distinguish between Instagram users with depression and healthy participants, just from viewing their photos. And while the answer was "yes" it turns out we're not as good as a computer program given the same task. 

Advertisement

"Obviously you know your friends better than a computer," co-author Chris Dansforth said in a statement. "But you might not, as a person casually flipping through Instagram, be as good at detecting depression as you think."

In yet another interesting finding, the computer algorithm was able to detect signs of depression before an individual received an official diagnosis, something the researchers believe has important implications.

"This could help you get to a doctor sooner," Danforth says. "Or, imagine that you can go to doctor and push a button to let an algorithm read your social media history as part of the exam."

Danforth acknowledges, however, that this certainly opens up a moral and ethical minefield - and that it's very early days. "We have a lot of thinking to do about the morality of machines," Danforth says. "So much is encoded in our digital footprint. Clever artificial intelligence will be able to find signals, especially for something like mental illness." 

In conjunction with a focus on upholding data privacy as well as "ethical analytics" the authors conclude: "The present work may serve as a blueprint for effective mental health screening in an increasingly digitalised society. More generally, these findings support the notion that major changes in individual psychology are transmitted in social media use, and can be identified via computational methods."