Through a filter darkly

On the heels of my recent post about the app Prisma, which renders photographs in the style of other artworks, I read an article that asserts that the filter you use on Instagram may be an indicator of depression

In a recent study, researchers Andrew Reece of Harvard University and Chris Danforth of University of Vermont built an algorithm which looked at patients’ Instagram feeds. They were able to diagnose depression with an accuracy rate of 70%. Their algorithm is better than a general practitioner’s assessment of depression after a face-to-face visit, which is only about 50% accurate.

There's a popular science fiction trope in which someone (or some entity) captures surprising patterns in the wealth of data that forms an online corona around each of us. It's not all that fictional; the technology exists right now, and is probably only limited by existing norms, and in certain places, legislation. Many people remember the consumer backlash when Amazon was caught altering prices based on user profiles; they dropped the practice, but so-called "dynamic pricing" is still in use elsewhere. 

All this had me thinking about pattern recognition in everyday life: how we all have areas of expertise in which an ostensibly-meaningless quantum of data (like an instagram filter) gives us unusual insight. One of the cool things about getting older is that I find my pattern-recognition is much more finely tuned than it was in earlier years. 

For example, my job brings me in contact with many people who are addicted to different drugs, particularly narcotics. I see patterns that clue me into a medical history of addiction much more quickly than would have been possible earlier in my career. I doubt I'm consciously aware of all the aspects of the pattern I'm perceiving - I just get a strong inclination, which is usually proven correct. 

I'm also, sadly, much more able to detect a pattern that says someone in front of me is about to die. You don't always get to choose the things you get good at.