Towards Digital Liminality: Computational tools for "beyond average" creative thinking

Renowned designer Kenya Hara writes: “Creativity is to discover a question that has never been asked.” It is in the early phases of the design process that these questions are discovered; an often ambiguous and liminal experience where new information is explored and considered in non-obvious ways to reveal unexpected meanings and associations. Digital technologies such as genetic algorithms and machine learning are increasingly employed in the tools used in these early phases of the design process. These computational techniques undoubtedly surpass humans at quickly generating a multitude of designs and calculating "optimized" responses, but their average-driven approaches are limited when it comes to embracing the serendipity that can often inspire creative breakthroughs. How can we develop digital tools to augment this liminal period of the creative process and help designers discover unexpected ideas?

This dissertation explores this question through three new "Beyond Average" systems that integrate ambiguity and serendipity into digitally enabled design tools: the Reframe creative prompt tool that juxtaposes language from a designer’s notes in surprising ways to provoke new associations between concepts in their project; the Looking Sideways inspiration exploration tool that presents a diverse range of unexpected content for each search query and suggests connections for the concepts discovered; and the digitally-augmented Design Daydreams ideation table and post-it note that seamlessly connects the physical and digital content that designers use in their creative process. These systems were informed by field research and interviews with expert designers and their impact on the design process was evaluated through several interventions in which creative practitioners, entrepreneurs and technologists used the Beyond Average tools to inspire new ideas for their projects. These interventions highlighted that the creative disruptions these tools provoke cannot exist alone; they must be situated in a larger design process that accommodates for serendipitous interjections and unexpected ideas. Overall, this research demonstrates how we can embed liminality into the digital tools we use to find inspiration as well as into the design process that helps us make use of these innovative ideas, and identifies new questions to consider as we design the future of our creative work.

Committee members:

Tod Machover
Muriel R. Cooper Professor of Music and Media & director of the Media Lab's Opera of the Future group,  Program in Media Arts and Sciences Academic Head
Massachusetts Institute of Technology

Daniel Cardoso Llach
Assistant Professor in the School of Architecture & co-director of the CoDe Lab
Carnegie Mellon University

Maria Yang
Associate Professor of Mechanical Engineering & D-Lab Faculty Academic Director
Massachusetts Institute of Technology