Social networks aspire to connect people, which is a noble but naive goal. When we uncritically accept connection as a good thing, we overlook difficult, important questions: Are some forms of virtual communication more nourishing than others? Might some in fact be harmful? Is it possible that Facebook, for instance, leaves some people feeling more lonely? No one knows for sure. We tend to build things first and worry about the effects they have on us later.
Robert Morris is taking the opposite approach. Starting with the desired effect of helping people deal with depression, he developed Panoply, a crowdsourced website for improving mental health. The site, which was the focus of his doctoral thesis at MIT Media Lab, trained users to reframe and reassess negative thoughts, embedding an established technique called cognitive behavioral therapy in an engaging, unthreatening interface. After a study confirmed the site's effectiveness, Morris formed a company and is now working on turning the idea into a polished consumer app.
Like other social networks, Panoply will take up that noble goal of connection, but in a more specific, structured way. As software goes, it's something of a novelty---a product that aims to enrich lives through precise, clinically-proven means, rather than merely assuming enrichment as a byproduct of its existence.
Morris, 34, studied psychology at Princeton under Nobel laureate Daniel Kahneman. A brief stint of clinical work after graduation left him impatient. "The efforts to adjust mental health were a little more incremental than I would've liked, especially when it came to using technology to address the problem," he says.
Though depression is thought to affect somewhere between 6 and 10 percent of US population, people avoid seeking treatment for many reasons. It can be expensive and inconvenient. It's also in many ways taboo. As a society, we don't make it easy for people to talk about mental health. As Morris sees it, technology has the potential to overcome all those barriers.
That belief led him to the Affective Computing group at MIT Media Lab, where Panoply was born, ironically, during a spell of low self-esteem. Upon arriving in Cambridge, Morris quickly came to feel like an imposter---he was a psychologist in a sea of highly capable programmers. "The expectation was that you could code at a world-class level," he says. "I felt very, very insecure."
Scrambling to advance his coding chops, Morris began spending time on Stack Overflow, a highly active programming community. It became an invaluable resource. Morris would ask elementary question about debugging code, and within minutes, someone would answer. "It was so amazing to have this crowd of people," he says. "It's just a marvel of crowdsourced intelligence."
The experience sparked a "sequence of intuitions" about his own work. Morris was familiar with the landscape of mental health software, including "teletherapy" apps that let users connect remotely with therapists and self-guided tools that include exercises users practice themselves. But teletherapy apps are expensive and often carry a whiff of stigma, and the self-guided resources are effective only if users can stay committed and "eat their broccoli," as Morris puts it.
Morris's love affair with Stack Overflow suggested another approach. He knew there were many simple techniques for helping people handle depression, and his coding community taught him how helpful it is having a group's support when learning a new skill. He remembers thinking: "Just like there are all these people helping me identify bugs in my code, perhaps I could create a similarly engaging and social system to help me identify bugs in my thinking."
If you've ever reassured a recently-dumped friend that there are plenty of fish in the sea, you've practiced a simple form of cognitive behavioral therapy. CBT is the blanket term for a number of techniques that help people identify negative thoughts and see them more objectively, a process often referred to as "reappraisal." As Morris puts it, "It's really about trying to readjust your thinking to bring better health."
Panoply was cleverly designed to help people internalize this skill. The site, which Morris built with a clinical psychologist from Northwestern University, invited anonymized users to describe a situation that was upsetting them. For example: "My roommate just came home, and I said 'Hi,' but he walked right by without looking at me." The app would then ask that user to write out interpretations of this event. They might say, "I don't think my roommate has ever liked me. I'm not popular. I'm not cool enough."
Posts like these triggered a three-tiered wave of crowdsourced action. The first person or two simply came by to lend support and sympathy (for his thesis study, Morris trained a group of workers from Mechanical Turk to pad the user base). A second wave read the entry and labeled specific places where the poster was distorting reality or thinking illogically. Then, a third group came and completely rewrote the initial story, casting the events in a less dire light. The system produced crowd-generated reappraisal unique to every dark thought. A professor of psychology from Stanford told MIT he thinks it's a "promising approach."
Crucially, though, Panoply was designed to work both ways. In addition to posting, users were encouraged to help give support, label bugs, and eventually rewrite entries themselves. In this way, the platform didn't just dispense reappraisal; it became a place to practice it.
Morris thinks this is vital. "You really learn these techniques best not by absorbing feedback from the crowd, but by teaching them to other people," he says. "Helping someone else reappraise a negative situation is a bite size task that only takes a couple of minutes, but you're really flexing that cognitive muscle over and over." As Morris sees it, if you can get people to practice the skill enough, it will eventually become second nature.
Morris' initial study on Panoply is being published this week in the Journal of Medical Internet Research. Compared to a control group that did a generic expressive writing exercise, users who initially exhibited signs of depression showed significant improvements for depression and reappraisal skills after using the platform for three weeks.
Recently, Morris formed a startup called Koko to turn Panoply into a consumer app. He's taking his time with it. For one, he's been thinking carefully about how to package the app---he doesn't want it to carry the stigmatized language of depression. He's also refining the three-step response, working to ensure the feedback system will be effective at scale. The challenge is to preserve the intended effect while borrowing from the interactive language of today's stickiest consumer apps. "We really want to make sure we're getting things right," he says.
It remains to be seen whether people will spend their precious time on a platform that could be seen as a repository of depressing anecdotes. But anonymous apps like Yik Yak and Whisper command our attention, and what's different here is that people are given a chance to actively help those strangers. Morris points out that many of the hired hands from Mechanical Turk asked to sign up for the site after the study was completed. He thinks the site could have mainstream appeal, pointing to the popularity of self-help books and all things related healthy living. "We know so much about how to adjust diet and how we eat, but we don't have a good framework collectively for our emotional health," he says. And he could be right. There are scores of social media platforms out there, but true happiness is a tremendously compelling feature.