About the Project
It has long been appreciated that humans form internal models that encode states of their environment and the relationship between those states. These representations are referred to as “cognitive maps” and are thought to provide a scaffold for adaptive behaviour, giving rise to higher cognitive functions like inference or knowledge transfer that have proven very challenging to reverse engineer in AI research.
Thus far, research has conceived cognitive maps as prospective representations that link states in their chronological order to predict future events, e.g., to decide which action will produce a desired outcome. However, recent work suggests that non-huma animals also use retrospective representations that link states in their reverse order to infer candidate causes of rare but significant outcomes. Prospective and retrospective maps are computationally distinct and may serve different roles in supporting decisions and learning. However, it is currently unknown how these types of maps are distinctly represented in the human brain, and how the information they provide is integrated for adaptive behaviour. For example, do retrospectively inferred causes serve to revise predictive models? And do prediction errors trigger retrospective information search? This project will tackle these questions via a multi-method computational neuroscience approach.
The PhD candidate will develop new tasks, and use cutting-edge neuroscience tools that measure (fMRI) and manipulate (TUS) human brain activity to probe how prospective and retrospective maps are uniquely encoded. Depending on the candidate’s interest, the project could also examine the role of prospective and retrospective maps in habit formation and extinction, which could have important implications in the context of addiction.
The candidate will receive in-depth training in computational modelling, neuroscience, and data analysis, allowing them to develop valuable skills for a career within or outside of academia.
Interested candidates are encouraged to get in contact (p.muhle-karbe@bham.ac.uk) for informal discussion before submitting an application.
Funding Notes
Fully funded PhD scholarship within the Midlands Integrative Biosciences Training Programme (MIBTP).
Details about the application procedure and assessment criteria can be found on the MIBTP website: View Website
References
- Behrens, T. E., Muller, T. H., Whittington, J. C., Mark, S., Baram, A. B., Stachenfeld, K. L., & Kurth-Nelson, Z. (2018). What is a cognitive map? Organizing knowledge for flexible behavior. Neuron, 100(2), 490-509.
- Jeong, H., Taylor, A., Floeder, J. R., Lohmann, M., Mihalas, S., Wu, B., & Namboodiri, V. M. K. (2022). Mesolimbic dopamine release conveys causal associations. Science, 378(6626), eabq6740.
- Luo, S., Li, M., Pan, D., Kingston, N., Klein-Flügge, M., Mahmoodi, A., & Rushworth, M. (2026). A frontal cortical world model shapes inference of causal relationships in hippocampus.
- Pan, D., D’Ambrogio, S., Kingston, N., Rascu, M., Sankhe, P., Luo, S., & Rushworth, M. F. (2025). Causal necessity of human hippocampus for structure-based inference in learning. bioRxiv, 2025-08.
- Muhle-Karbe, P. S., Sheahan, H., Pezzulo, G., Spiers, H. J., Chien, S., Schuck, N. W., & Summerfield, C. (2023). Goal-seeking compresses neural codes for space in the human hippocampus and orbitofrontal cortex. Neuron, 111(23), 3885-3899.
- Namboodiri, V. M. K., & Stuber, G. D. (2021). The learning of prospective and retrospective cognitive maps within neural circuits. Neuron, 109(22), 3552-3575.