You probably already know that excessive social media time can be bad for your mental health.
A new study suggests that your social media behavior can be a surprisingly reliable indicator or your mental health.
Researchers from Harvard and the University of Vermont devised a computer program that analyzes the frequency of posts—along with several different criteria like dominant color and subject matter—to see patterns in the posting habits of people with diagnosed mental health issues, specifically depression. They used the data gathered to see how people with depression use the platform differently than the rest of the population.
Published by EJP Data Science, the study determined that people with depression were more likely to post more frequently than others, and tended to rack up more comments. They were also less likely to use filters, but when they did use filters, the images posted tended to be darker while containing more blue and gray hues. Users affected by depression were also more likely to post images including faces (though interestingly, their images tended to include fewer faces per photo than those of non-depressed users).
These trends seem to match up with existing literature on depression. Dr. Christopher Danforth, co-director of the University of Vermont’s Computational Story Lab and one of the study’s authors, told BuzzFeed News, “It seems to be the case that people who are experiencing depression see the world in a darker, bluer fashion quite literally, and they spend less time in social groups.”
The study monitored 166 users, and who collectively posted a total of 43,950 images.
Approximately half of the users studied had been diagnosed with depression in the three years preceding the study.
The algorithm the researchers developed showed surprising effectiveness, correctly identifying users dealing with depression 70 percent of the time. While that may not sound especially impressive at first, previous research has shown that general practitioners are only able to correctly identify depressed patients 42 percent of the time.
While the relative success of the researchers’ algorithm in identifying depressed users is certainly promising, there’s much more work to do before implementation on a wider scale. Because the sample size was relatively small, there’s still a chance that the results won’t successfully scale up.
As Danforth told BuzzFeed News, there’s “no guarantee that this would translate to the average Instagram user.” He emphasized that this is simply a proof of concept, not a diagnostic test.
When it comes to potential uses, Danforth had a few ideas.
He suggested that the system could be used to help mental health professionals identify those more likely to experience depression, suggesting that the technology could even have the potential to save lives.
Danforth shared, “Imagine an app you can install on your phone that pings your doctor for a
Technology could play a major role in our understanding of mental health. Someday, your psychiatrist may check your Instagram—and according to Danforth, that could be a good thing.
