Investigating properties of functional brain networks using the resonance phenomenon

UoM administered thesis: Phd


“Investigating properties of functional brain networks using the resonance phenomenon.” A thesis submitted to The University of Manchester for the degree of Doctor of Philosophy by Caroline Lea-Carnall (2017). Neural oscillations are ubiquitous in the brain and are thought to underlie many of our cognitive capabilities. However, although oscillations occur within a wide frequency range, different brain regions exhibit resonance-like characteristics at specific points in the spectrum. At the microscopic scale, it is well documented that individual neurons exhibit resonance characteristics, indicating that this is an intrinsic feature of neural circuitry. However, questions remain as to whether cortical resonance is a direct result of the oscillatory activity of individual neurons or an emergent property of the networks that interconnect them and if so, whether we can exploit this feature of the brain to measure or modulate network connectivity. In Chapter 3 we characterise the effects of network topology on resonance frequency using computational modelling and a human electroencephalography (EEG) experiment. We show that cortical resonance depends not only on the resonance of the individual oscillators but also on network properties such as network size and connection density. In Chapter 4, we use a combination of neurocomputational modelling and functional magnetic resonance imaging (fMRI) to show that repetitive tactile stimulation applied to the digits at or away from the resonance frequency of the human somatosensory system (SI) has distinct outcomes in terms of behavioural and physiological measures of connectivity. We found that stimulation of just 45 mins produced changes in cortical topography measurable by fMRI, including a shifting of the digit regions following above-resonance co-stimulation only. These results were further evidenced by computational modelling work. Finally, in Chapter 5, we describe the results of a functional magnetic resonance spectroscopy experiment and show that it is possible to modulate gamma-aminobutyric acid (GABA) concentrations, an inhibitory neurotransmitter thought to play a role in plasticity, in human SI using periodic stimuli presented at two driving frequencies, at- and above-resonance of SI. We found that GABA levels within SI decreased by more than 50% after co-stimulation with above-resonance frequency only. We also found that performance on a psychophysics task was impaired after above-resonance co-stimulation only and that the degree of impairment was significantly correlated to the decrease in GABA levels. In conclusion, computational modelling of neural networks is proposed to be invaluable as a tool to integrate the experimental, clinical and theoretical neurosciences. The specific studies described in this thesis contribute to our understanding of connectivity and plasticity in healthy human subjects and make testable predictions for use in clinical applications.


Original languageEnglish
Awarding Institution
Award date31 Dec 2017