Among the findings: depressed individuals posted with a greater frequency and received more comments on their posts (but fewer likes), and also were more likely to post photos of faces, yet had a "lower average face-count per post" compared to the healthy participants. Those without depression "disproportionately favored the Valencia filter, which lightens the tint of photos". (Roughly half of participants had reportedly suffered from clinical depression within the past three years).
Prof Danforth and U.S. colleague Andrew Reece from Harvard University wrote in a blog post accompanying the study: "Pixel analysis of the photos in our dataset revealed that depressed individuals in our sample tended to post photos that were, on average, bluer, darker, and greyer than those posted by healthy individuals".
Depression is strongly associated with reduced social activity [20, 21].
The study's tool detected a few key common traits among depressed people's feeds, most of which had to do with color schemes: They more frequently used blue, grey, and dark tones, and generally avoided filters, but when they did use them, they predominantly preferred the black-and-white Inkwell option.
The team found that people with depression did indeed post photos that were, on average, bluer, darker, and grayer. Yet, those volunteers were not able to complete the task as efficiently as the statistical computer model; the rating done by humans lacked much correlation with the features of the photos that were identified by the computer.
"There are reasons why depression is called blue, and why people associate red with raging, and why people say depression is like a dark or black cloud", Galynker said.
The researchers were then able to predict which photos showed signs of depression using the study mode.
US Not Seeking to Topple North Korea Regime, Says Tillerson
Meanwhile, a senior Republican senator said President Trump considered going to war with North Korea as an option. He also said President Trump promised him "if there's going to be a war to stop them, it will be over there".
The methodology used meant that people with depression were correctly identified 70 percent of the time.
As part of the study, published in the journal EPJ Data Science, volunteers attempted to distinguish between Instagram photos posted by depressed and healthy people.
And using these factors, they were able to nearly double the likelihood of correctly diagnosing a patient with the technology, compared to a doctor in a consultation.
"It's a proof of concept and for the particular individuals we studied, this set of predictors works for them".
"We have a lot of thinking to do about the morality of machines", he said. Ideally, Danforth says the best outcome of technology like this is getting those individuals the medical support that they need.
More than 500 participants initially were recruited for the study, the researchers noted, but many dropped out because they would not consent to sharing their social media data.