Medical or psychiatric help has become quite expensive services these days. However, people could skip costly routine checks and diagnose their own mental state according to new research. All they have to do is take a look at their Instagram history of photos to diagnose their own mental state according to new research.
Researchers Analyzed Data from 166 Volunteers with Instagram Accounts
Researchers of a new study claimed that the way people choose to organize their social media profiles reflects better their personality than they might think. The lead authors of the research are Andrew Reece of Harvard University and Chris Danforth, a professor at the University of Vermont’s Department of Mathematics & Statistics. Together they gathered 166 participants. They were asked to share their Instagram feeds for scientific reasons together with their history of mental health.
Around half of volunteers had at least one diagnosis of depression over the last three years. The next step was to run the collected data through a statistical computer model. The goal of the software was to identify any visual signs of depressive feelings.
The computer algorithm received as many such markers as possible that were backed by science. For instance, a study implied that people who are feeling down choose photos for social media that are grayer, bluer, and darker than usual.
The Research Points out that Social Media Photos Reflect the Mental State of Users
The findings of this study point out that such markers might be in alignment with the real mental state of the owners of the photos. Another example is that people with depression tend to avoid filters. However, in case that they are going for one they are choosing on average the Inkwell filter. This effect renders the pictures in black and white colors only.
Another marker of depression is when users post photos with faces such as selfies. However, on average the number of persons in the photos is smaller than the ones that healthy people post. This might mean that they interact with fewer people than the others. All in all, the computer model managed to diagnose the state of Instagram users at a percentage of 70% correctly.
On the other hand, the authors of the research aren’t proposing this computer model as a screening tool just yet. They need extensive studies to reach the level of a standalone diagnostic test.
Image source: 1