Bachelor / Master Thesis

This thesis will explore the interaction between causal machine learning (Causal ML) models – like causal large language models – and human decision-making in collaborative contexts. The primary focus will be on understanding how causal insights provided by AI systems impact human reliance, potentially resulting in over-reliance (blind trust) or under-reliance (excessive skepticism). The research will begin with a literature review (e.g, systematic or scoping review) to synthesize current knowledge and identify gaps in the field. Building on this foundation, the thesis will conceptualize potential research avenues to address these gaps. One selected research question will be examined in depth through the design, execution, and analysis of an experiment. Here, the identification and development of suitable ML-task for the experiment is one central aspect of the thesis. The findings will assess the effects of causal insights on decision quality across critical domains such as healthcare, finance, and education, ultimately providing actionable recommendations for developing AI systems that promote balanced and effective human-AI collaboration. 

Recommended introductory literature: 

Zheng, M., Marsh, J. K., Nickerson, J. V., & Kleinberg, S. (2020). How causal information affects decisions. In Cognitive Research: Principles and Implications (Vol. 5, Issue 1). Springer Science and Business Media LLC. https://doi.org/10.1186/s41235-020-0206-z  

Kleinberg, S., Marsh, J.K. Less is more: information needs, information wants, and what makes causal models useful. Cogn. Research 8, 57 (2023). https://doi.org/10.1186/s41235-023-00509-7 

Max Schemmer, Niklas Kuehl, Carina Benz, Andrea Bartos, and Gerhard Satzger. 2023. Appropriate Reliance on AI Advice: Conceptualization and the Effect of Explanations. In Proceedings of the 28th International Conference on Intelligent User Interfaces (IUI '23). Association for Computing Machinery, New York, NY, USA, 410–422. https://doi.org/10.1145/3581641.3584066 

Hemmer, P., Schemmer, M., Kühl, N., Vössing, M., & Satzger, G. (2024). *Complementarity in Human-AI Collaboration: Concept, Sources, and Evidence*. arXiv. https://arxiv.org/abs/2404.00029