Dissertation Title: Designing for Deep Engagement
Abstract:
Flow state represents the quality of meaningful experience-- an effortless, depth of attention that is often undermined in our interrupt-driven, modern society. In this thesis, I present four novel interventions to promote the occurrence of these deep cognitive states, including one I tested thoroughly on myself for a month.
Evaluating whether one of these interventions has a meaningful impact on flow state is difficult to do. The bulk of my work, then, focuses on the methodological challenges of flow state research. Herein I tackle three weaknesses in our ability to make strong, generalizable predictions about the causal link between environmental stimuli and flow states: (1) I discuss advancing how we represent the environment (specifically for aural stimuli) using phenomenological principles; (2) I advance the state-of-the-art in how we represent and measure flow bio-behaviorally (with the goal of integrating physiology into our judgements); and (3) I evaluate methodological weaknesses in current experimental flow work. To do this, I present experimental work on models of auditory attention, new wearables and survey instruments for flow estimation, and an experiment that compares flow as measured in lab and at home across varying task structures.
Armed with tools for naturalistic data collection and time-aware, probabilistic representations of cognition, this work improves the way we make generalizable claims about the influence of design on flow states. In so doing, it points to an improved approach for social psychology more generally. Our statistical tools need re-grounding in intuition; our latent, cognitive ontological constructs demand stochastic representations; and our models require idiographic structures informed by longitudinal, naturalistic data.
Committee members:
Joe Paradiso, Alexander W. Dreyfoos Professor of Media Arts and Sciences, MIT
Rosalind Picard, Professor of Media Arts and Sciences, MIT
Gloria Mark, Professor of Informatics, University of California, Irvine
Jan-Willem van de Meent, Professor of Computer Science, University of Amsterdam and Northeastern University