A vector-field paradigm for the human encephalogram and its application to Alzheimer’s Disease
File(s)
Author(s)
Vinao Carl, Matteo L.
Type
Thesis
Abstract
Alzheimer’s disease (AD) is marked by the progressive accumulation of amyloid-β and tau pathology alongside widespread disruption of cortical network dynamics. While abnormal large-scale activity contributes to both symptom expression and disease progression, approaches for restoring healthy brain-wide coordination remain limited. Here, I tested whether global network dynamics can be modulated using phase-locked stimulation (PLS), which synchronizes sensory input to the phase of ongoing cortical oscillations. In contrast to prior correlational findings, causal PLS experiments in healthy participants revealed no consistent frequency- or phase-specific effects on attention, and reanalyses suggested earlier effects may reflect methodological artefacts. These results challenge the notion that attentional function has a fixed oscillatory signature or can be localised to a single brain region, instead pointing to a distributed neural code spanning across multiple spatial and temporal scales.
To unveil this multi-scale architecture I developed a novel framework that applies divergence and curl operators to instantaneous phase-gradient fields, producing time-resolved maps of cortical information flow. Eigen-decomposition of these flow fields revealed a low-dimensional set of conserved routing modes—spanning individuals and frequency bands—that structure the propagation of neural activity across the cortex. In AD, impaired flexibility in transitioning between these modes (“rigid routing”) tracked regional atrophy and predicted deficits in attention and memory. Together, these findings identify a core organizing principle of cortical communication, introduce a biomarker of cognitive dysfunction, and suggest new targets for network-level neuromodulation in neurodegenerative disease.
To unveil this multi-scale architecture I developed a novel framework that applies divergence and curl operators to instantaneous phase-gradient fields, producing time-resolved maps of cortical information flow. Eigen-decomposition of these flow fields revealed a low-dimensional set of conserved routing modes—spanning individuals and frequency bands—that structure the propagation of neural activity across the cortex. In AD, impaired flexibility in transitioning between these modes (“rigid routing”) tracked regional atrophy and predicted deficits in attention and memory. Together, these findings identify a core organizing principle of cortical communication, introduce a biomarker of cognitive dysfunction, and suggest new targets for network-level neuromodulation in neurodegenerative disease.
Version
Open Access
Date Issued
2024-07-01
Date Awarded
01/08/2025
License URL
Advisor
Grossman, Nir
Hampshire, Adam
Sharp, David
Sponsor
Engineering and Physical Sciences Research Council
Grant Number
EP/L016737/1
Publisher Department
Department of Brain Sciences
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)