PerDIS 2026 Closing Keynote

We invite participants of the German Pre-CHI Event to joint the PerDIS 2026 Closing Keynote.

Keynote: Roel Vertegaal – Beyond Attentive Displays: From Attention-Aware to Surprise-Aware Design

Professor Dr. Roel Vertegaal is Chair of Human-Computer Interaction (HCI) at Radboud University in Nijmegen, the Netherlands, where he studies, models, and builds user experiences of tomorrow. He is the founding director of the Human Media Lab, known for pioneering attentive user interfaces — including the attention-aware techniques now common in most smartphones — as well as organic user interfaces such as the foldable phone. His work spans a broad range of empirical, theoretical and design contributions to HCI for which he was elected a member of ACM SIGCHI Academy. His current research focuses on neuromorphic design, with particular interest in (inter)active inference.

Abstract: Over the past 25 years, attention-aware displays have gone from a niche research idea to wide-spread adoption. Originally invented to combat information overload in a future world of many devices, attention-aware displays are more relevant than ever in a world of fragmented attention spans. Attentive UX is now everywhere: in cars, smartphones, ambient displays, virtual reality headsets, and smart homes. In this talk, I revisit what attention actually is, why eye tracking does not describe it sufficiently, and how surprise forms a new design commodity for pervasive displays. Surprise is what results from the user’s brain predicting the world – a discrepancy between prediction and observation, in bits. It builds in the brain as a potential energy that is released as action, learning or emotion. Within this neuromorphic design framework, user performance and error can be analyzed through a single formalism. The talk closes with some examples of how surprise-aware displays might be a solution for keeping humans in the loop with AI agents, and an agenda for the design and evaluation of surprise-aware interaction using methods borrowed directly from statistical mechanics.