Oh, Hyerin
(2024)
Investigating metabolite and functional responses in working memory using magnetic resonance techniques.
PhD thesis, University of Nottingham.
Abstract
Working memory, or the ability to temporarily maintain and manipulate received information, is critical for other cognitive functions such as attention. Since working memory function deficits can be found in different brain disorders, it is essential to understand the physiological processes underpinning working memory. This thesis aims to understand the neural basis of working memory, focusing on metabolite and functional responses using functional magnetic resonance spectroscopy (fMRS) and functional magnetic resonance imaging (fMRI) using N-back tasks.
First, metabolic and functional responses to working memory in the dorsolateral prefrontal cortex (DLPFC) were investigated. Increased unresolved glutamate and glutamine (Glx) levels during working memory, along with anti-correlations between Glx changes, working memory, blood oxygen level-dependent (BOLD) signals and working memory performance, were observed. In metabolic processes, glucose is processed to pyruvate through glycolysis which then enters in the TCA cycle by processing into acetyl-CoA. α-ketoglutarate, which can interchange with glutamate, is generated in the TCA cycle. The known task-induced increase in cerebral glucose metabolism and potential increase in the transformation of α- ketoglutarate, a key TCA intermediary, to glutamate suggests that the observed Glx increase is a marker of upregulated TCA activity supporting working memory. Anti-correlations between working memory performance and neural activity changes, as well as between working memory performance and the response to neural activity, may support the neural efficiency hypothesis. This hypothesis suggests that individuals with better working memory require less neural activity.
Secondly, an analysis of functional changes during working memory using high-quality human connectome project (HCP, n = 125) database showed sensorimotor network (SMN) deactivation in addition to the activation of the core network of working memory (frontoparietal network, FPN). Generally elevated between network functional connectivity (FC) with several decreased between network FC (across sensorimotor, visual, dorsal attention networks) was also observed. Decreased SMN activity with increased FPN activity during working memory suggested a decoupling between SMN and FPN. While FC with FPN increased, reduced FC with SMN might support the idea of functional decoupling between FPN and SMN. Generally enhanced network interactions might be required for working memory processing, and some decreased interactions might be balanced with generally increased interactions across networks.
Extended from the study with HCP dataset, positive correlations between working memory performance and FC at rest and in working memory were found. These findings may contribute to future studies (e.g., both inter- network FC involving visual working memory at rest and during working memory can predict working memory performance by showing positive correlations) investigating associations between FC and working memory performance.
This thesis highlights a general neurometabolic model of task-induced increases in energy metabolism and enhanced neural network communications across different networks in working memory.
Actions (Archive Staff Only)
 |
Edit View |
Tools
Tools
![[thumbnail of Corrected_Thesis_HO_281024.pdf]](https://eprints.nottingham.ac.uk/style/images/fileicons/application_pdf.png "Corrected_Thesis_HO_281024.pdf")