Computational Decision Making and Learning Group

Head: Nicolas Schuck

The mission of the Computational Decision Making and Learning Group is to understand (1) replay and representation learning in the brain, (2) the role of cognitive maps in spatial and non-spatial decision-making and (3) how ageing and psychiatric diseases affect these processes.

In our replay research, we have developed an important methodological foundation for studying fast sequential activations in the human brain using fMRI (Wittkuhn et al, Nature Comms, 2021). Studying replay with fMRI offers the important advantage that it allows to investigate the full anatomical distribution of replay events, and to shed light on the interplay, or independence, of hippocampal and extrahippocampal replay events during rest. In line with this idea, we discovered hippocampus independent replay in visual cortex following a non-mnemonic task (Witttkuhn et al. Nature Comms, 2021). We have also begun to connect the theme of replay with the theme of representation learning in an extensive review of computational literature of replay (Wittkuhn et al., Neuroscience and Biobehavioral Reviews, 2021) and, in a collaboration with Ray Dolan, argued that spontaneous neural activity offers important window into cognition in general (Liu et al., Nature Reviews Neuroscience, 2022).

In work related to decision making and cognitive maps, we have shown how humans use relational mental codes to generalise in non-spatial environments (Wu et al., PLoS Computational Biology, 2020), and explored foraging in fixed size environments (Hall-McMaster et al., iScience, 2021).

In studies related to psychiatry and ageing, we have investigated how anxiety affects computational mechanisms of learning (Zika et al., under review/bioRxiv) and shown how ageing and dopamine affect spatial direction signals in the brain (Koch et al, Neuropsychologia, 2020; Koch et al, bioRxiv/under review, 2022).

The group has continued to grow (two new postdocs and one PhD student joined), attract further funding (two new grants awarded by the German Research Association (DFG) to PI Nicolas Schuck, and 5 competitive PhD or Postdoc scholarships were to members of the group). Of particular relevance is one grant which funds a collaboration with another group from the MPS-UCL center, Quentin Huys’ Applied Computational Psychiatry Group, and Mingbo Cai (U Tokyo), in which mind wandering and rumination in depressive patients will be studied.

Key publications

1. Liu, Y., Nour, M. M., Schuck, N. W., Behrens, T. E., & Dolan, R. J. (2022). Decoding cognition from spontaneous neural activity. Nature Reviews Neuroscience, 23(4), 204-214.
2. Wittkuhn, L., & Schuck, N. W. (2021). Dynamics of fMRI patterns reflect sub-second activation sequences and reveal replay in human visual cortex. Nature communications, 12(1), 1-22.
3. Wu, C. M., Schulz, E., Garvert, M. M., Meder, B., & Schuck, N. W. (2020). Similarities and differences in spatial and non-spatial cognitive maps. PLoS computational biology, 16(9), e1008149.