The composition of images posted on popular social media platforms might serve as indicators of depression, suggest researchers at the University of Vermont.
Chris Danforth and Andrew Reece sourced three years’ worth of Instagram feeds of 166 people – some 43,950 images in all – along with each person’s mental health history. Participants were selected so that roughly half in the study had experienced episodes of depression. {%recommended 1365%}
Their images and data were then fed into a computer, with an algorithm grading the images according to factors including colour density, composition and presence of faces.
In their paper published in the journal EPJ Data Science, Danforth and Reece report the algorithm, using clinically verified insights from psychological practice, correctly identified the participants with a history of depression in 70% of cases. This compares, they say, with 40% correct in-person diagnosis of depression in the US.
Despite this finding, though, the scientists are quick to point out it does not yet constitute a reliable indicator of mental health. “This study is not yet a diagnostic test, not by a long shot,” Danforth says, “but it is a proof of concept of a new way to help people.”
The study’s key finding is that depressed people were more likely to favour images steeped in blue, grey and generally dark tones. They were also more likely to use monochrome filters. “In other words, people suffering from depression were more likely to favor a filter that literally drained all the color out the images they wanted to share,” the report concludes.
On the basis of images posted, the computer was in several cases able to reach a diagnosis earlier than the date at which individuals consulted and were diagnosed by a doctor. Though more research is still needed, Danforth says this aspect is particularly encouraging. “This could help you get to a doctor sooner,” he 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 examination.”
