Research: Fachbereich Humanwissenschaften

Research directions

Human cognition is remarkably flexible. We seemingly adapt our perception, thinking, and behavior effortlessly to changing circumstances, pursue (and discard) self‑set goals, and take our own uncertainty into account when making decisions and acting. This level of flexibility has not yet been achieved in artificial intelligence.

Our research investigates the cognitive mechanisms underlying this flexibility and how they are implemented in the brain. We are also interested in how these processes contribute to mental health, and how they develop across the lifespan.

Approach

Our approach for studying these questions can be summarised as measure, model, perturb.

Measure: We develop innovative cognitive tasks for both adults as well as children that allow us to measure the behavioural and neural signatures (using EEG) of the cognitive process of interest reliably and efficiently.

Model: We develop generative models of behaviour and neural activity that reflect our mechanistic understanding of these cognitive processes, building upon Bayesian and reinforcement learning modelling frameworks.

Perturb: While participants perform our tasks, we interfere with specific neural systems to test their causal role in supporting the cognitive process. For this we use pharmacology (targeting specific receptor types in the brain) and non-invasive brain stimulation with focussed ultrasound (targeting specific brain regions or nuclei).

Current projects

The neurocognitive mechanisms of repetitive negative thoughts (Wellcome Trust Mental Health Award, Feb 2026 - Jan 2031)

In collaboration with researchers at the University of Oxford (lead PI: Prof. Mike Browning), ETH Zürich and the University of Exeter, we are launching a £5 million project beginning in February 2026. Using a combination of MRI, non‑invasive brain‑stimulation (transcranial magnetic stimulation and transcranial ultrasound stimulation), computational modelling and a novel pupil‑based biofeedback paradigm, the project aims to uncover the neurocognitive mechanisms that make repetitive negative thoughts—or rumination—start and persist, and to test whether pairing brain stimulation with a simple cognitive‑training exercise can reduce these patterns. More information can be found at the Oxford news page.

Relevant publications:

Algermissen, J., Rascu, M., Weber, L. A., Boer, T. den, Martin, E., Treeby, B., Gray, M., Cleveland, R. O., Wittmann, M. K., Clarke, W. T., & others. (2025). Low-intensity focused ultrasound to human amygdala reveals a causal role in ambiguous emotion processing and alters local and network-level activity. bioRxiv, 2025–08.

Studying flexible decision making and belief updating in changing environments using continuous cognitive tasks and hierarchical Bayesian models

Humans constantly adapt their behaviour to changes in their circumstances, which occur over many timescales, from milliseconds to years. We want to understand the cognitive mechanisms that allow us to achieve this flexibility. Using novel continuous decision making tasks, we study how participants integrate information across different timescales and how they continuously adapt their learning and decision strategies to changes in their environment. We optimize task design to efficiently and reliably measure the behavioural and neural signatures (using M/EEG) of these processes, while developing cognitive modelling frameworks to describe the underlying cognitive computations.

Relevant publications:

Ruesseler*, M., Weber*, L. A., Marshall, T. R., O’Reilly, J., & Hunt, L. T. (2023). Quantifying decision-making in dynamic, continuously evolving environments. eLife, 12, e82823.
Weber, L. A., Waade, P. T., Legrand, N., Møller, A. H., Stephan, K. E., & Mathys, C. (2023). The generalized hierarchical Gaussian filter. arXiv Preprint arXiv:2305.10937.
Foucault, C., Weber, L. A., & Hunt, L. T. (2025). Environmental dynamics shape human learning: Change points versus random walks. bioRxiv, 2025–11.
Legrand, N., Weber, L., Waade, P. T., Daugaard, A. H. M., Khodadadi, M., Mikuš, N., & Mathys, C. (2024). pyhgf: A neural network library for predictive coding. arXiv Preprint arXiv:2410.09206.
Mathys, C., & Weber, L. (2020). Hierarchical Gaussian filtering of sufficient statistic time series for active inference. Communications in Computer and Information Science, 1326, 52–58.

Rethinking reinforcement learning: the interoceptive origins of reward

We all love a good ice-cream. But what exactly is rewarding about consuming it? In conventional reinforcement learning models, the environment emits scalar ‘ground-truth’ reward signals that the agent can use to learn what to do. But in biological agents, ‘reward’ is subjective, dynamic and state-dependent – generated within the organism, and inferred from noisy interoceptive signals. In the Cognitive Modelling group, we study the subjectivity and flexibility of reward functions in biological agents using both experiments and conceptual work extending conventional RL models. We are also interested in how this perspective can inform our understanding of disturbances in reward learning across mental health.

Relevant publications:

Weber, L. A., Yee, D. M., Small, D. M., & Petzschner, F. H. (2025). The interoceptive origin of reinforcement learning. Trends in Cognitive Sciences, 29(9), 840-854.
Müller-Schrader, M., Petzschner, F. H., Heinzle, J., Kasper, L., Wellstein, K. V., Bayer, J., Engel, M., Bianchi, S., Weber, L. A., Pruessmann, K. P., & others. (2025). Differences in layer-specific activation of the insula during interoceptive vs. exteroceptive attention. bioRxiv, 2025–10.
Petzschner, F. H., Weber, L. A. E., Gard, T., & Stephan, K. E. (2017). Computational Psychosomatics and Computational Psychiatry: Toward a Joint Framework for Differential Diagnosis. Biological Psychiatry, 82(6), 421–430.
Petzschner*, F. H., Weber*, L. A., Wellstein, K. V., Paolini, G., Do, C. T., & Stephan, K. E. (2019). Focus of attention modulates the heartbeat evoked potential. NeuroImage, 186, 595–606.
Stephan, K. E., Manjaly, Z. M., Mathys, C. D., Weber, L. A. E. E., Paliwal, S., Gard, T., Tittgemeyer, M., Fleming, S. M., Haker, H., Seth, A. K., & Petzschner, F. H. (2016). Allostatic Self-efficacy: A Metacognitive Theory of Dyshomeostasis-Induced Fatigue and Depression. Frontiers in Human Neuroscience, 10, 550.

The dual role of ketamine: a computational perspective on pharmacological interventions in psychiatry

In psychiatry, the NMDA receptor antagonist ketamine takes on two very different roles: a pharmacological model of psychosis, due to its psychotomimetic effects, and a potent, fast-acting antidepressant. We want to understand whether ketamine produces its effects in these different domains - perceptual (psychosis) and affective (depression) - via a common neurocognitive mechanism. This question is crucial for drug development, and for our understanding of disease mechanisms in psychosis and depression. In an ongoing study with collaborators at the University of Oxford, we use novel continuous decision tasks, EEG, and cognitive modelling to study the neurocognitive mechanisms by which ketamine affects perception and affective processing.

Relevant publications:

Weber*, L. A., Diaconescu*, A. O., Mathys, C., Schmidt, A., Kometer, M., Vollenweider, F., & Stephan, K. E. (2020). Ketamine Affects Prediction Errors about Statistical Regularities: A Computational Single-Trial Analysis of the Mismatch Negativity. Journal of Neuroscience, 40(29), 5658–5668.
Weber, L. A., Tomiello, S., Schöbi, D., Wellstein, K. V., Mueller, D., Iglesias, S., & Stephan, K. E. (2022). Auditory mismatch responses are differentially sensitive to changes in muscarinic acetylcholine versus dopamine receptor function. eLife, 11.