Are You Depressed? Check Your Social Media Photos

Your social media photos are probably full of grins, gorgeous vacation destinations, your kids at their cutest and your dog at his cutest. Or perhaps you capture everyday moments—a pretty sunrise, a farmer’s market, a nice view from your office window. It turns out, those photos might be a good predictor of your mental health, —but maybe not in the way you think.

We’re all more likely to post photos of happy moments rather than, say, our faces mid-migraine or a short video of our last argument with our partner. But, if a new study of Instagram photos is anything to go by, we’re still letting on more than we think we are. Researchers at Harvard University and the University of Vermont studied 43,950 photos from 166 volunteers. The participants answered questions about their social media habits and history of depression diagnosis, and then given a clinical depression survey. Their Instagram photos, meanwhile, were analyzed by hue, brightness, color saturation, Instagram filter, post frequency and more.

The depressed participants had a few things in common when it came to their Instagram photos:

- They were more likely to post photos with faces, but the photos had fewer faces than photos from non-depressed Instagram users. “Fewer faces may be an oblique indicator that depressed users interact in smaller social settings, which would be in accordance with research linking depression to reduced social interactivity,” the researchers write.

 - They used more filters overall and were partial to the black and white Inkwell filter, while non-depressed participants favored Valencia. This is in line with other research that’s found a correlation between depression and a preference for darker colors.

- They posted more frequently. Earlier this year, a study from the University of Pittsburgh Schools of the Health Sciences found that young adults who checked their social media sites more often than their peers were 2.7 times more likely to have depression.

So you have permission to start diagnosing your Instagram friends from behind your smartphone screen, right? Not exactly (well, not at all, actually). But the findings are promising in terms of how machine learning techniques can help screen for depression. According to the researchers, more than half of general practitioners’ depression diagnoses were false positives, which is costly for both healthcare programs and patients. The majority of the Instagram algorithm’s depression classifications were correct. “These findings suggest new avenues for early screening and detection of mental illness,” the researchers suggest.

This kind of data can also be useful for the social media sites themselves—blogging platform Tumblr is already doing it with keywords, serving up a message of concern and list of helpful resources when someone searches words like “depressed” and “suicidal.”

117 comments

John B
John B11 days ago

Interesting. Thanks for sharing.

SEND
Sarah Hill
Sarah Hill7 months ago

thanks

SEND
Jim Ven
Jim Ven7 months ago

thanks for sharing.

SEND
federico bortoletto

Grazie della condivisione.

SEND
Jim Ven
Jim Ven8 months ago

thanks for the article.

SEND
Nathan D.
Nathan D8 months ago

Dissatisfaction with where we are in life is a great motivator. If we were happy all the time there would be no need or room for improvement.

SEND
Siyus Copetallus
Siyus Copetallus8 months ago

Thank you for sharing.

SEND
Philip Watling
Philip Watling9 months ago

I was hit by a car in December 1994 and the old me stopped... My life was taken from me! I did not turn to drugs, cigarettes, alcohol - I never once got depressed. Why not? Read my inspiring book Flight of a Lifetime to find out. Find it on Facebook :)

SEND
chris B.
chris B9 months ago

More depressing seeing everyone else traveling and having fun. No thanks. I'm fine where I am.

SEND
Quanta Kiran
Quanta Kiran9 months ago

noted

SEND