Research and Innovation Aim

Research & Innovation

Bringing AI-based systems into real-world use in ways that are truly useful in practice, trustworthy in sensitive domains, and tailored to individual use cases as well as the individuals using them.

Research and innovation at different scales (from individual research, to independent research institute leadership, to operational and strategic large-scale initiative leadership) following three threads: Human-AI Interaction, Digital Health & Prevention, and Personalization & Adaptive Systems. These threads run through publications, funded projects, teaching, and the current chapter with the Bavarian AI Foundation Model Initiative alike.

A current synthesis of this agenda is my Habilitation (under review, LMU Munich): it introduces health making as the ongoing cultivation of health through human-technology entanglements that extend across time and life circumstances, and develops principles and architectures for Precision Adaptive Behavior Change (PABC), resting on three interdependent foundations: adaptive system architectures, human-centred behaviour change design, and context representation through personal health data. The work is grounded empirically with a main application area in cardiovascular prevention and rehabilitation, and points to human-AI relations in health, encompassing aspects of accountability, trust calibration, and governance of life-accompanying technologies, as structurally required considerations and key focus areas for future research and innovation.

Human-AI Interaction

Making AI systems usable and grounded in real-world interactions: from early work on embodied language acquisition at Sony CSL to conversational interfaces, human-centred evaluation of AI systems, and the question of how people and learning systems can work together productively and accountably.

The Last JITAI? Exploring Large Language Models for Issuing Just-in-Time Adaptive Interventions: Fostering Physical Activity in a Prospective Cardiac Rehabilitation Setting

D. Haag, D. Kumar, S. Gruber, D. P. Hofer … J. David Smeddinck · Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems

Bot or not? User Perceptions of Player Substitution with Deep Player Behavior Models

J. Pfau, J. David Smeddinck, I. Bikas, R. Malaka · Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems

Digital Health & Prevention

Technology-assisted behaviour change and digitally supported patient pathways, with a focus on cardiac rehabilitation and prevention — co-designed with clinicians and patients, and evaluated in real care settings including randomised controlled trials.

mHealth Support in Cardiac Care Pathways for Patient Self-Management During Transitions From Hospital to Rehabilitation: Exploratory Field Study

I. Höppchen, S. Tino Kulnik, A. Meschtscherjakov, J. Niebauer … D. Wurhofer · JMIR Cardio

Smartphone Apps for Food Purchase Choices: Scoping Review of Designs, Opportunities, and Challenges

R. Benthem de Grave, C. N. Bull, D. Monjardino de Souza Monteiro, E. Margariti … J. David Smeddinck · Journal of Medical Internet Research

Personalization & Adaptive Systems

Adaptivity as a first-class design material: adaptive difficulty in motion-based games for health, just-in-time adaptive interventions, and LLM-driven personalisation that adapts support to individual context while remaining transparent and controllable.

Personality as Relational Infrastructure: User Perceptions of Personality-Trait-Infused LLM Messaging

D. P. Hofer, D. Haag, R. Islambouli, J. David Smeddinck · Preprint

How to Present Game Difficulty Choices?: Exploring the Impact on Player Experience

J. D. Smeddinck, R. L. Mandryk, M. V. Birk, K. M. Gerling, D. Barsilowski, R. Malaka · Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems

Effects of Balancing for Physical Abilities on Player Performance, Experience and Self-esteem in Exergames

K. Maria Gerling, M. Miller, R. L. Mandryk, M. Valentin Birk, J. David Smeddinck · Proceedings of the SIGCHI Conference on Human Factors in Computing Systems

Full publication list →  ·  All projects →

Current chapter: contributing my discipline-linking perspective to a Bavaria-wide effort, long in the making and carried by researchers across eleven universities: supporting the development of open, multimodal AI foundation models that bring AI into real applications, for science, public administration, and the people they serve.