Crowdsourcing mental health and emotional well-being

Abstract

More than 30 million adults in the United States suffer from depression. Many more meet the diagnostic criteria for an anxiety disorder. Psychotherapies like cognitive-behavioral therapy can be effective for conditions such as anxiety and depression, but the demand for these treatments exceeds the resources available. To reach the widest possible audience, mental health interventions need to be inexpensive, anonymous, always available, and, ideally, delivered in a way that delights and engages the user. Towards this end, I present Panoply, an online intervention that administers emotion- regulatory support anytime, anywhere. In lieu of direct clinician oversight, Panoply coordinates support from crowd workers and unpaid volunteers, all of whom are trained on demand, as needed. Panoply incorporates recent advances in crowdsourcing and human computation to ensure that feedback is timely and vetted for quality. The therapeutic approach behind this system is inspired by research from the fields of emotion regulation, cognitive neuroscience, and clinical psychology, and hinges primarily on the concept of cognitive reappraisal. Crowds are recruited to help users think more flexibly and objectively about stressful events. A three-week randomized controlled trial with 166 participants compared Panoply to an active control task (online expressive writing). Panoply conferred greater or equal benefits for nearly every therapeutic outcome measure. Statistically significant differences between the treatment and control groups were strongest when baseline depression and reappraisal scores were factored into the analyses. Panoply also significantly outperformed the control task on all measures of engagement (with large effect sizes observed for both behavioral and self-report measures). This dissertation offers a novel approach to computer-based psychotherapy, one that is optimized for accessibility, engagement and therapeutic efficacy.