Chatbot / Experiment
9 out of 12 participant would use a digital therapist again
During the summer of 2011, I worked at the Calming Technology Lab at Stanford. We were looking for ways to use technology in a manner that induces cognitive and physiological calm. To me, this was revolutionary because the common perception is that technology distracts us. My hypothesis was that we could help decrease daily stresses through therapeutic and rational vocalisations.
POM was a simple natural language processing script that I pulled from a NLTK tutorial. There were 12 participants in my experiment and every day they would receive a text asking if they were stressed. If yes, then the bot would message (vague) question and statements in an attempt to get the participant to wonder. The experiment ran for 3 months, and we qualitatively measured success by asking “Do you feel less stressed now?” at the end of each session.