The research group "Human-Centric Explainable AI", led by Dr. David Johnson, is directly related to the TRR in its focus on co-construction. The group aims to develop theories on how people process explanations they receive from AI-assisted decision-making systems in high-stakes situations, such as mental health assessment, and to work out which types of explanations and interactions are best suited to reduce overreliance on AI recommendations that are not trustworthy or accurate. The goals are to improve the design of XAI-based tools and interactions by involving users in the design process, and to conduct extensive evaluations to better understand how people use explanations. With this knowledge, new and improved interactive AI-assisted decision-making systems for mental-health assessment and other real-world problems will be developed. To ensure that the research covers all important aspects to meet these goals, the group takes an interdisciplinary approach involving computer science, psychology, and human-computer interaction.