“The collaboration with Cristina Conati enriches our research spectrum in a special way,” emphasizes Prof. Dr. Elisabeth André. “She brings a unique combination of AI methodological expertise and a human-centered perspective - ideal for our interdisciplinary focus.”

The stay is funded by the Alexander von Humboldt Foundation, whose research award recognizes top international researchers for their scientific achievements to date and enables them to establish long-term collaborations with research institutions in Germany.

Personalized explainability instead of a black box

Cristina Conati is one of the leading international experts in the field of artificial intelligence, human-computer interaction (HCI) and cognitive science. Her work focuses on the question of how AI systems can be designed in such a way that they take into account the cognitive states, needs and expectations of their users - while remaining transparent and comprehensible

A particular focus lies on the research field of explainable AI. The aim is to develop systems that are not only technically efficient, but also create trust by revealing their decision-making processes. An earlier stay at Prof. Dr. André's chair in 2023 led to a joint publication on the so-called Alterfactual method - a procedure that shows how decisions are made by AI systems and which features are irrelevant to the AI's decision making. The communication of irrelevant features is particularly important when people want to know whether the AI might be biased and, for example, discriminates against individuals in decision-making processes based on their ethnic background.

Diverse fields of application - from education to medicine

During her current research stay in Augsburg, Prof. Dr. Conati is working together with Prof. Dr. André's team on the application of these concepts in various socially relevant contexts. These include the analysis of eye movement data as an indicator of cognitive learning processes, the development of AI-supported diagnostics in medicine and adaptive training systems in sport.

Methodologically, it combines data-based procedures with psychological findings on human learning and behavior. The goal: AI systems that not only inform, but also provide individual support - and do so in a way that remains comprehensible to users.

Scientific excellence with international appeal

Cristina Conati is Professor of Computer Science at the 伟德国际_伟德国际1946$娱乐app游戏 of British Columbia and is active in various international professional societies. Among other things, she was president of the HUMAINE Association for Research on Emotion in Human-Computer Interaction and program and conference chair of the renowned ACM Conference on Intelligent User Interfaces (IUI). She is a Fellow of AAAI, the Association for the Advancement of AI, and an Editor in Chief for the Journal of AI in Education. Her contributions to user-centred AI have been recognised with numerous awards, including the Test of Time Award from the Educational Data Mining Society.

Her stay in Augsburg not only stands for a close scientific cooperation, but also for research that places people at the center of technical developments. The joint projects with Prof. Dr. André's chair contribute to further expanding the interface between technology, cognition and social relevance.

tb