Uploaded February 2026 | Updated September 2026, 8 hours ago
Artificial intelligence is transforming scientificdiscovery across disciplines—from predicting protein structures in biology toidentifying novel quantum states in physics. In the natural sciences, automateddiscovery systems now enable researchers to navigate vast experimental andhypothesis spaces, moving beyond the limits of human intuition. In this talk, Iexplore how these methods can be adapted to the study of mind and behavior,introducing automated scientific discovery as a paradigm for cognitive science.I present AutoRA, an open-source framework for automating key stages ofempirical research—including experimental design, data collection, and modelinference—and demonstrate how it can be used to uncover new computationalmodels of cognition and novel behavioral phenomena. Through case studies inpsychophysics, learning, decision-making, and cognitive control, I illustratehow closed-loop discovery systems can support more integrative, scalableapproaches to understanding human cognition. Finally, I discuss the uniquechallenges that automated scientific discovery faces in cognitive science, suchas experimental fragmentation and the need for multi-level explanations, andoutline future directions for building systems that not only accelerate scientificprogress, but reshape how we generate and evaluate theories of the mind.
Sebastian Musslick is an Assistant Professor of Computational Neuroscience at Osnabrück University and a Visiting Faculty Member at the Department for Cognitive and Psychological Sciences at Brown University. He is leading the Laboratory for Automated Scientific Discovery of Mind and Brain at the Institute of Cognitive Science at Osnabrück University, and is directing the Autonomous Empirical Research Group.
Artificial intelligence is transforming scientificdiscovery across disciplines—from predicting protein structures in biology toidentifying novel quantum states in physics. In the natural sciences, automateddiscovery systems now enable researchers to navigate vast experimental andhypothesis spaces, moving beyond the limits of human intuition. In this talk, Iexplore how these methods can be adapted to the study of mind and behavior,introducing automated scientific discovery as a paradigm for cognitive science.I present AutoRA, an open-source framework for automating key stages ofempirical research—including experimental design, data collection, and modelinference—and demonstrate how it can be used to uncover new computationalmodels of cognition and novel behavioral phenomena. Through case studies inpsychophysics, learning, decision-making, and cognitive control, I illustratehow closed-loop discovery systems can support more integrative, scalableapproaches to understanding human cognition. Finally, I discuss the uniquechallenges that automated scientific discovery faces in cognitive science, suchas experimental fragmentation and the need for multi-level explanations, andoutline future directions for building systems that not only accelerate scientificprogress, but reshape how we generate and evaluate theories of the mind.
Sebastian Musslick is an Assistant Professor of Computational Neuroscience at Osnabrück University and a Visiting Faculty Member at the Department for Cognitive and Psychological Sciences at Brown University. He is leading the Laboratory for Automated Scientific Discovery of Mind and Brain at the Institute of Cognitive Science at Osnabrück University, and is directing the Autonomous Empirical Research Group.
