Core Cognition

The Max Planck Research Group Core Cognition is working on a unified theory of human cognition. To this end, the team members are developing computer-based cognitive models that translate existing psychological theories into algorithms. These methods are intended to investigate the brain’s information-processing mechanisms that cannot be directly inferred from observable behavior.

Since such models have so far been limited to individual tasks or research areas, the research team aims to develop large-scale, integrated models of human cognition. These models are intended to simulate and predict human behavior across a variety of experimental tasks, while also providing comprehensible explanations for the underlying processes involved in the emergence of human behavior.

Three challenges must be addressed to achieve this:

• The Technical challenge: How do we even create such algorithmic models?
• The scientific challenge: How do we use existing knowledge to gain new insights into human cognition?
• The applied challenge: How can these insights be translated into real-world applications?

Methodologically, approaches from cognitive science, machine learning, and language modeling are employed. These include, amongst others, studies of human conversations as well as complex visual and computer-based tasks. In the long term, the insights gained here are expected to pave the way for new applications in education and psychiatry.

Key Publications:

Binz, M., Akata, E., Bethge, M., Brändle, F., Callaway, F., Coda-Forno, J., ... & Schulz, E. (2025). A foundation model to predict and capture human cognition. Nature.

Binz, M., & Schulz, E. (2023). Using cognitive psychology to understand GPT-3. Proceedings of the National Academy of Sciences.

Binz, M., Alaniz, S., Roskies, A., Aczel, B., Bergstrom, C. T., Allen, C., ... & Schulz, E. (2025). How should the advancement of large language models affect the practice of science? Proceedings of the National Academy of Sciences.