Translations
Abstract
This thesis builds on a body of sociotechnical research at the MIT Center for Constructive Communication that draws upon "ancient wisdoms" of dialogue and listening and harnesses the power of technology to inform the design of dialogue spaces that promote deep, meaningful, and authentic conversations. Our approach hinges on the belief that society functions best when we hear and understand each other, an outcome that our work strives to advance by exposing people to the personal stories of others in ways that connect rather than divide. I take inspiration from anthropological practices and recent Data Humanism and Activism epistemologies to derive a set of design considerations for restorative interfaces. These principles inform the development of Translations, an interactive experience that invites audiences to more deeply engage with a curated collection of stories surfaced during small group facilitated conversations. The design of this visual and auditory experience allows audiences to explore stories they may otherwise not hear through websites that center thematic summaries and high level insight visualizations. The selected stories are curated using AI emotion analysis and sensemaking which are leveraged to draw the user’s attention to moments of interest across conversations, such as moments of affirmation. The efficacy of this curation method to engender empathy and emotional disruption, precursors to restorative listening, is evaluated and the results from user tests for and interviews about the overarching interface are discussed. Ultimately, this thesis presents both a framework for automatic curation of audio narratives as well as an interactive interface to present these selected stories, both of which have wide-ranging applications in the media and civic space.
This interactive website was implemented as part of my Master's Thesis at the MIT Media Lab.
