Keynote Talk 1 (Wed, 7 Oct. 2026)
Jenny Spurlock | Senior Product Design Director, Meta Reality Lab
Title: The Art and Science of Interaction Pathfinding
Abstract
Interaction defines a computing platform much more strongly than a display; it is the ultimate reconciliation of human capabilities, technology capabilities, and use cases, connecting humans to the things they want to do. For spatial computing platforms, defining that interaction model is a fundamentally different problem than refining one: we must pathfind through vast, under-explored design spaces where technologies, human capabilities, and use cases have not yet converged. This act of pathfinding is both an art and a science. Drawing on nine years of leading input and interaction pathfinding at Meta Reality Labs—from VR controllers, eye, and hand tracking to AR glasses, neural interfaces, and AI—this talk explores key methods and mental models for navigating defining an interaction model for a new platform. Through real-world examples, the talk will cover a range of ideas: (1) using strategic prototyping to map unknown design spaces; (2) solving second- and third-order problems to make interaction feel truly delightful; (3) establishing frameworks and numerical floors for “good enough” probabilistic input; (4) applying principles from historical platform transitions to guide future input evolution; and more. These unique approaches form a virtuous cycle between prototyping, testing, and theorizing that begins as breadth-first exploration and ends in confident product convergence in zero-to-one spaces.
Speaker Biography
Jenny Spurlock is a Senior Product Design Director at Meta Reality Labs, where she leads input & interaction pathfinding to help define how people will interact with and control experiences across Meta’s virtual and augmented reality products. Since joining Meta in 2017, she has led various multi-disciplinary design, hardware, software, and research science teams dedicated to zero-to-one interaction innovation whose pathfinding efforts have shaped every input modality and interaction model Meta has shipped (and invented many that haven’t). This includes eye and hand tracking, controller concept development, surface input, neural band, and AI interaction. Through the sisyphean process of years of iteration and working with novel and ambiguous technologies over and over again, she has developed a unique process that combines prototyping, design taste, conceptual models, and blending disciplines to unlock the future for human-computer interaction in spatial computing.
Keynote Talk 2 (Thu, 8 Oct. 2026)
Antonio Rizzo | Full Professor of Cognitive Science and Technology, University of Siena
Title: What Sailing Can Teach Us About Augmented Agency
Abstract
Why can an experienced sailor coordinate changing wind, hull motion, rope tension, and the invisible geometry of a course with practiced fluency, while a simple menu in mixed reality can still become cognitively demanding? This keynote uses sailing not only as a metaphor, but as a design lens for understanding skilled cognition in complex environments. Sailing reveals that intelligent action does not arise from commands alone. It emerges through a continuous coupling among perception, anticipation, bodily movement, tools, cultural practices, other people, and the environment. Drawing on Helmholtz’s account of perception as unconscious inference, William James’s analysis of habit and sensorimotor continuity, Vygotsky’s theory of cultural mediation, and ecological and distributed approaches to cognition, the talk examines what these traditions can teach us about spatial interaction. Future augmented and mixed reality systems should move beyond placing digital information in front of the eyes. They should help people read situations, coordinate action, and develop competence without weakening situational awareness or taking control away from them. Physical anchors, haptic cues, gaze-sensitive interaction, and spatially aware artificial agents can become parts of such a system, provided that their interventions remain timely, legible, reversible, and negotiable. Augmented reality can then be understood as Augmented Intelligence: not an invisible pilot, but a cognitive crew member that helps people anticipate, decide, learn, and act while leaving them firmly at the helm.
Speaker Biography
Antonio Rizzo is Full Professor at the University of Siena, where he works at the intersection of cognitive science, human-computer interaction, and artificial intelligence. His research focuses on human-centered design, cognitive ergonomics, and the use of intelligent technologies in industry, healthcare, and complex socio-technical systems. He has served as Chair of the European Association of Cognitive Ergonomics, as a member of the Scientific Committee Initiative de Recherche sur l’Éducation et la Formation of the French Government, and as a member of the NATO WG30 Human Factors and Human Reliability Group. From 1995 to 1997, he acted as Apple Inc. Liaison for the dissemination of the User-Centered Design process within the Apple Design Project. Beyond academia, he has worked with companies including Philips Design, Siemens, FRANKE, Trenitalia, and MPS. He also co-designed UDOO, a single-board computer integrating microcontroller and microprocessor capabilities.
Keynote Talk 3 (Fri, 9 Oct. 2026)
Misha Sra | Associate Professor, University of California, Santa Barbara
Title: Spatial Human-AI Interaction
Abstract
Most interactions with AI happen through language using prompts, instructions, and explanations in a chat box. But in many everyday activities, language alone is not enough. People rely on maps, diagrams, environments, and their own bodies to ground reasoning and coordinate action. To work with people in these settings, AI must perceive the same context, track it as it changes, and express intent back into it.
These requirements form the three pillars of our proposed framework for human–AI coordination. Shared grounding establishes a spatial context that both people and AI can perceive and refer to. Temporal continuity maintains that shared understanding as the activity unfolds. Intent expression makes each party’s intended actions legible within the context. Together, these pillars define Spatial Human–AI Interaction, a paradigm in which the “system” is no longer the technology alone; it is the human–AI loop.
In the talk, I will present the framework and research that develops new coordination methods, identifies design principles, and evaluates spatial human–AI systems in real-world settings. I will close by introducing XARP (XR Agent-ready Remote Procedures), a toolkit that bridges the XR and AI development ecosystems through a shared Python interface for both developers and AI agents.
Speaker Biography
Misha Sra, PhD, is an Associate Professor at UC Santa Barbara and the founding director of the Human-AI Experience (HAX) Lab. Her research focuses on spatial human–AI interaction, exploring how people and AI systems coordinate through shared physical environments and spatial representations, including maps, layouts, and game boards. Her interdisciplinary lab designs, builds, and evaluates interaction techniques, software toolkits, human–AI systems, hardware platforms, sensing pipelines, and datasets to study how AI affects human learning, performance, and behavior. This work aims to inform the design of newer human–AI systems that can effectively augment a wide range of daily human activities.
Sra’s research has been published in top tier journals and conference venues for XR, HCI, and AI, including IEEE VR, ISMAR, VRST, ACM CHI, IUI, DIS, AHs, COLM, NAACL, ICLR, EMNLP, and Nature Machine Intelligence. Her work has been featured in MIT News, Forbes, The Verge, Engadget, PCMag, Santa Barbara Independent, and more. She received an NSF CAREER Award for her research on human–AI interaction (2023–2028), the IEEE VGTC Best Dissertation Award (2020), a Silver Edison Award for design, and recognition as an MIT Rising Star in Engineering (2018). Her collaborators have included Toyota North America, Snap Inc., Samsung, Google, MIT, Stanford University, and the University of Tokyo, among others.