Human-Centered Design

The choice of modality and of the moment of presentation is governed today largely by
convention and by the constraints of whatever hardware is at hand, and only rarely by
evidence about the user, the task and the situation. This matters most where the
situation is restrictive: hands occupied, sterile field, vehicle in motion, limited
connectivity, users without technical training who are working from inferences they
have no way of verifying.

We develop computational models of perception, attention and cognitive load that
predict which form of presentation supports which task under which constraint, and we
derive design rules for adaptive multimodal presentation from them, evaluated in
physical, extended and virtual realities. Where alternatives exist we compare them
against each other: contact-free input, augmented overlays, projection onto the work
surface and an ordinary screen each fail differently, and which failure is tolerable
is a property of the task rather than of the technology.

In collaborative projects this is usually our contribution rather than the project's
goal, which puts us in a position to watch a great many systems fail for reasons that
have nothing to do with their models. We therefore design with the people who will use
a system rather than for them, and we work with social scientists where the question
runs past the interface — with police sociology on how de-escalation is practised and
taught, with agricultural sociology on why recommendations are or are not acted upon.
Most of the tools of digital agriculture already exist, and the open question is why
they are not adopted: whether a recommendation is understood, whether it can be
contested, and whether it fits the way decisions are actually taken.

P. Chojecki, D. Przewozny, F. Tirschmann, F. Schmid, U. Jerzembeck, D. Runde, K. Brukamp, H. Aubin, and S. Bosse. ‘A-PLR: An Activity-Aware Assistant for Accurate and Usable Passive Leg Raise Tests in the Smart ICU’. in: Proceedings of the 31st International Conference on Intelligent User Interfaces. IUI ’26: 31st International Conference on Intelligent User Interfaces. Paphos Cyprus: ACM, Mar. 23, 2026, pp. 928–940

M. T. Lafci, B. Nierula, D. Damar, and S. Bosse. ‘Too Close for Comfort? Investigating Virtual Professor Distance and Student Learning in VR’. in: 2025 IEEE International Conference on Systems, Man, and Cybernetics (SMC). 2025, pp. 1052–1057

D. Przewozny, D. Runde, N. Hoerner, N. Gard, S. Bosse, and P. Chojecki. ‘Comparison of Uni- and Multimodal Interfaces for Spatial Interaction’. In: Proceedings of the Symposium on Spatial User Interaction. ACM. Online, CA, USA: ACM, 2022, pp. 1–2

P. Chojecki, D. Strazdas, D. Przewozny, N. Gard, D. Runde, N. Hoerner, A. Al-Hamadi, P. Eisert, and S. Bosse. ‘Assessing the Value of Multimodal Interfaces: A Study on Human–Machine Interaction in Weld Inspection Workstations’. In: Sensors 23 (2023), p. 5043