Title: Understanding Individual, Stakeholders, and AI's Roles in Navigating Through Life-Changing Events
Date: August 21st (Friday)
Time: 12pm - 2 pm EST
Location: GVU Cafe (TSRB 2nd Floor)
Kefan Xu
Ph.D. Student in CS (Human-Computer Interaction)
School of Interactive Computing
Georgia Institute of Technology
Committee Members
Dr. Rosa I. Arriaga (Advisor) - School of Interactive Computing, Georgia Institute of Technology
Dr. Alex T. Adams - School of Interactive Computing, Georgia Institute of Technology
Dr. Jennifer G. Kim - School of Interactive Computing, Georgia Institute of Technology
Dr. Mark W. Newman - School of Information, University of Michigan
Dr. Andrea G. Parker - School of Interactive Computing, Georgia Institute of Technology
Abstract
Life-changing events (LCEs)—health issues, job changes, shifts in the caregiving network—disrupt routines, relationships, and identities. LCEs pose an acute challenge in individuals’ lives. They disrupt not only the individuals’ established practices but also ripple through the broader situated ecology, affecting friends, families, and the relationships among them. Yet existing HCI research has largely examined LCEs at a single scale, without accounting for how they propagate across dyadic and ecological layers.
This dissertation investigates LCEs at three scales. At the individual level, I developed Planneregy, a mobile app supporting reflective iteration on physical activity routines; a 42-day deployment study showed how bundling plans into strategies and scaffolding weekly reflection help individuals adapt to shifting circumstances. At the dyadic level, analyzing informal caregiving discussions on Reddit, I characterized how LCEs escalate caregiving conflicts and how online health communities support caregivers' sense-making during transitions. At the ecological level, I interviewed 22 diabetes patients to map how LCEs reshape the care ecology. Based on these insights, I proposed ecological informatics as a design guideline for technologies that support information work across ecological layers.
Building on these insights, my proposed work develops EcoCare, an AI-assisted visualization system that helps patients, caregivers, and clinicians see how LCEs impact interconnected ecological entities. Using EcoCare as a probe in a mixed-methods study, I will investigate how stakeholders make sense of LCEs and make decisions when they can see the full ecological landscape. This will serve to evaluate the ecological informatics framework. Collectively, this dissertation advances an ecological understanding of LCEs and contributes guidelines for designing systems that support people navigating major life changes—alone, with intimate others, or within the broader care ecology.