Social Judo Reports on Parents in the dark

The answer is it certainly looks that way, according to research done by Andrew Reece at Harvard University and Chris Danforth at the University of Vermont. They have found significant connections between the colors used on Instagram posts and an individual’s mental health status. Yes, you read that correctly, color and mental health.

In fact, in their research they reported that the link was so strong that it could actually be used as a tool to detect mental illnesses like depression. Reece and Danforth used 170 Instagram users, of whom 70 were clinically depressed. They collected photos from before their diagnosis with depression and then asked the non depressed users to rate them. Using this information, they were then able to create a scientific system to spot the connections between depressed mood and photographic images. They then used their system and got it right 54% of the time. Compared with physicians who correctly diagnose depression 42% of the time, Reece says “It’s not an A+, but it’s a 25% improvement over those human rates.”

Research on the connection between color and mood tell us that the darker and grayer colors are commonly associated with darker or more depressed mood, and lighter colors indicate lighter mood or perhaps a more cheerful disposition. So here is what Reece and Danforth found in their study on Instagram and depression:

“Using only photographic details, such as color and brightness, our statistical model was able to predict which study participants suffered from depression, and performed better than the rate at which unassisted general practitioners typically perform during in-person patient assessment”.

● Depressed people posted images that were darker, grayer and bluer and they received fewer “likes” than people who were not depressed.

● Participants who were not depressed more often choose to lighten their photos using the filter called Valencia.

● Users without depression got more likes while users with depression got more comments.

Although their research has not yet been peer reviewed and they agree that more research needs to be done, Reece and Danforth say “[t]hese findings support the notion that major changes in individual psychology are transmitted in social media use, and can be identified via computational methods.” Now, can Twitter posts tell us if someone is depressed? One study found that not only can it tell but it may even be able to predict it. A team of researchers said their model could scan and detect depression among Twitter users with an accuracy rate of 70%.

Microsoft Research followed 476 Twitter users, 171 of whom were depressed. They searched the tweets of users up to one year before they were diagnosed with depression for language, level of engagement, mentions of certain types of medications and other criteria. They then compared this information with non depressed Twitter users and came up with a model to predict depression. Based on over 115 earlier studies, the team determined that general practitioners only accurately diagnose depression about 42% of the time, their model could predict depression 70% of the time.

The accurate diagnosis of mental health conditions of course requires a much more in depth look than just the analysis of Instagram filters and Tweets, but if used as a part of the entire tool kit, it might just help some get help much earlier.

We are just beginning to examine the ways social media can help us to detect depression but it is interesting to think that maybe one day, our social media use will be an early detection tool for all kinds of mental illness.

Andrea DiFilippo, MSW, LICSW