Neural Manifold Approaches to Understanding Memory and Cognitive Change in Dementia

Supervisory Team:

• Prof. Simon Schultz — Department of Bioengineering, Imperial College London
• Dr JeYoung Jung — School of Psychology, University of Nottingham

Based at: Imperial College London (with collaborative visits to the University of Nottingham)

Funding: Fully-funded (UK), stipend £23,805 p.a., duration 3.5 years

Start date: From 1st January 2027

Application deadline: Oct 31st, 2026

Project Description

Dementia is characterised by progressive, often devastating, changes to memory and cognition – yet the underlying neural mechanisms driving these changes remain poorly understood at the level of large-scale brain dynamics. In recent years, neural manifold approaches – techniques that describe how the collective activity of large neural populations can be captured by low-dimensional, structured geometric representations have been of great interest for understanding of how the brain encodes, transforms, and retrieves information. To date, these approaches have been developed and validated primarily in animal models, including in prior work from the Schultz laboratory at Imperial College London, where manifold-based analyses have revealed how neural population dynamics relate to memory function and its disruption in models of neurodegenerative disease.

In this studentship you will work on the translation and adaptation of these neural manifold techniques for the analysis of human neuroimaging data, in order to study how memory-related neural representations are altered in dementia. Working jointly with Dr JeYoung Jung's group in the Department of Psychology at the University of Nottingham – which brings expertise in human cognitive neuroscience and neuroimaging of memory and ageing – the successful candidate will develop novel computational methods for extracting and interpreting low-dimensional neural manifolds from human neuroimaging datasets (e.g. fMRI data from the PREVENT, with potential additional access to magnetic resonance spectroscopy data). We will apply these methods to investigate how memory-related neural geometry changes across healthy ageing and in dementia.

The project offers an exciting opportunity to work at the interface of computational neuroscience, engineering, and clinical cognitive neuroscience, contributing new analytical tools with the potential to advance understanding of, and ultimately biomarkers for, cognitive decline in dementia.

Key research questions the project may address:

• How can neural manifold analysis techniques developed for animal population recordings be adapted to the constraints and structure of human neuroimaging data?
• Do memory-related neural manifolds show measurable geometric changes with healthy ageing versus in dementia?
• Can manifold-based features provide more sensitive markers of early cognitive decline than conventional neuroimaging analyses?

Candidate Requirements

We are seeking a highly motivated candidate with:

• A strong background (BSc/MSc/MEng, or equivalent) in engineering, computer science, physics, mathematics, or another computational/quantitative discipline.
• Strong programming and quantitative/computational analysis skills.
• An interest in, and ideally some prior experience with, the analysis of human neuroimaging data (e.g., fMRI, EEG, or MEG).
• Some prior knowledge of, or coursework in, neuroscience.
• Enthusiasm for interdisciplinary research spanning computational methods development and cognitive/clinical neuroscience.

Prior experience with dimensionality reduction techniques, machine learning, or population-level neural data analysis (e.g., in Python or MATLAB) would be an advantage but is not essential.

Training and Environment

The student will be based in the Schultz Laboratory, Department of Bioengineering, Imperial College London, a leading centre for computational and systems neuroscience research, with regular collaborative engagement with the Jung Laboratory in the School of Psychology at the University of Nottingham. The student will benefit from training in cutting-edge computational methods, human neuroimaging techniques, and the opportunity to work across two thriving research environments and institutions.

How to Apply

Please email Prof. Schultz at the address below with CV, a brief cover letter describing your motivation for wanting to carry out the PhD and transcripts. References will be requested from shortlisted applicants.

For informal enquiries about the project, please contact:

• Prof. Simon Schultz: s.schultz@imperial.ac.uk
• Dr JeYoung Jung: jeyoung.jung@nottingham.ac.uk

Imperial College London and the University of Nottingham are committed to equality, diversity and inclusion, and welcome applications from all suitably qualified candidates.

Funding Notes

Funding: Fully-funded (UK), stipend £23,805 p.a., duration 3.5 years